The rapid evolution of artificial intelligence, especially in large language models (LLMs), has significantly impacted various domains, including healthcare. In chest X-ray (CXR) analysis, previous studies have employed LLMs, but with limitations: either underutilizing the LLMs' capability for multitask learning or lacking clinical accuracy. This article presents M4CXR, a multimodal LLM designed to enhance CXR interpretation. The model is trained on a visual instruction-following dataset that integrates various task-specific datasets in a conversational format. As a result, the model supports multiple tasks such as medical report generation (MRG), visual grounding, and visual question answering (VQA). M4CXR achieves state-of-the-art clinical accuracy in MRG by employing a chain-of-thought (CoT) prompting strategy, in which it identifies findings in CXR images and subsequently generates corresponding reports. The model is adaptable to various MRG scenarios depending on the available inputs, such as single-image, multiimage, and multistudy contexts. In addition to MRG, M4CXR performs visual grounding at a level comparable to specialized models and demonstrates outstanding performance in VQA. Both quantitative and qualitative assessments reveal M4CXR's versatility in MRG, visual grounding, and VQA, while consistently maintaining clinical accuracy.

The purpose of this study is the development of a deep learning model based on nnU-Net v2 for the automated segmentation of sphenoid sinus and middle skull base anatomic structures in cone-beam computed tomography (CBCT) volumes, followed by an evaluation of the model's performance. In this retrospective study, the sphenoid sinus and surrounding anatomical structures in 99 CBCT scans were annotated using web-based labeling software. Model training was conducted using the nnU-Net v2 deep learning model with a learning rate of 0.01 for 1000 epochs. The performance of the model in automatically segmenting these anatomical structures in CBCT scans was evaluated using a series of metrics, including accuracy, precision, recall, dice coefficient (DC), 95% Hausdorff distance (95% HD), intersection on union (IoU), and AUC. The developed deep learning model demonstrated a high level of success in segmenting sphenoid sinus, foramen rotundum, and Vidian canal. Upon evaluation of the DC values, it was observed that the model demonstrated the highest degree of ability to segment the sphenoid sinus, with a DC value of 0.96. The nnU-Net v2-based deep learning model achieved high segmentation performance for the sphenoid sinus, foramen rotundum, and Vidian canal within the middle skull base, with the highest DC observed for the sphenoid sinus (DC: 0.96). However, the model demonstrated limited performance in segmenting other foramina of the middle skull base, indicating the need for further optimization for these structures.

This study explored the effects of using Intelligent Quick Magnetic Resonance (IQMR) image post-processing on image quality in Field of View Optimized and Constrained Single-Shot Diffusion-Weighted Imaging (FOCUS-DWI) sequences for prostate cancer detection, and assessed its efficacy in distinguishing malignant from benign lesions. The clinical data and MRI images from 62 patients with prostate masses (31 benign and 31 malignant) were retrospectively analyzed. Axial T2-weighted imaging with fat saturation (T2WI-FS) and FOCUS-DWI sequences were acquired, and the FOCUS-DWI images were processed using the IQMR post-processing system to generate IQMR-FOCUS-DWI images. Two independent radiologists undertook subjective scoring, grading using the Prostate Imaging Reporting and Data System (PI-RADS), diagnosis of benign and malignant lesions, and diagnostic confidence scoring for images from the FOCUS-DWI and IQMR-FOCUS-DWI sequences. Additionally, quantitative analyses, specifically, the peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM), were conducted using T2WI-FS as the reference standard. The apparent diffusion coefficients (ADCs) of malignant and benign lesions were compared between the two imaging sequences. Spearman correlation coefficients were calculated to evaluate the associations between diagnostic confidence scores and diagnostic accuracy rates of the two sequence groups, as well as between the ADC values of malignant lesions and Gleason grading in the two sequence groups. Receiver operating characteristic (ROC) curves were utilized to assess the efficacy of ADC in distinguishing lesions. The qualitative analysis revealed that IQMR-FOCUS-DWI images showed significantly better noise suppression, reduced geometric distortion, and enhanced overall quality relative to the FOCUS-DWI images (P < 0.001). There was no significant difference in the PI-RADS scores between IQMR-FOCUS-DWI and FOCUS-DWI images (P = 0.0875), while the diagnostic confidence scores of IQMR-FOCUS-DWI sequences were markedly higher than those of FOCUS-DWI sequences (P = 0.0002). The diagnostic results of the FOCUS-DWI sequences for benign and malignant prostate lesions were consistent with those of the pathological results (P < 0.05), as were those of the IQMR-FOCUS-DWI sequences (P < 0.05). The quantitative analysis indicated that the PSNR, SSIM, and ADC values were markedly greater in IQMR-FOCUS-DWI images relative to FOCUS-DWI images (P < 0.01). In both imaging sequences, benign lesions exhibited ADC values markedly greater than those of malignant lesions (P < 0.001). The diagnostic confidence scores of both groups of sequences were significantly positively correlated with the diagnostic accuracy rate. In malignant lesions, the ADC values of the FOCUS-DWI sequences showed moderate negative correlations with the Gleason grading, while the ADC values of the IQMR-FOCUS-DWI sequences were strongly negatively associated with the Gleason grading. ROC curves indicated the superior diagnostic performance of IQMR-FOCUS-DWI (AUC = 0.941) compared to FOCUS-DWI (AUC = 0.832) for differentiating prostate lesions (P = 0.0487). IQMR-FOCUS-DWI significantly enhances image quality and improves diagnostic accuracy for benign and malignant prostate lesions compared to conventional FOCUS-DWI.

Liver diseases pose a significant global health challenge due to their silent progression and high mortality. Proper interpretation of radiology reports is essential for the evaluation and management of these conditions but is limited by variability in reporting styles and the complexity of unstructured medical language. In this context, Natural Language Processing (NLP) techniques and Large Language Models (LLMs) have emerged as promising tools to extract relevant clinical information from unstructured liver radiology reports. This work reviews, from a practical point of view, the current state of NLP and LLM applications for liver disease classification, clinical feature extraction, diagnostic support, and staging from reports. It also discusses existing limitations, such as the need for high-quality annotated data, lack of explainability, and challenges in clinical integration. With responsible and validated implementation, these technologies have the potential to transform liver clinical management by enabling faster and more accurate diagnoses and optimizing radiology workflows, ultimately improving patient care in liver diseases.

The hippocampal-to-ventricle ratio (HVR) is a biomarker of medial temporal atrophy, particularly useful in the assessment of neurodegeneration in diseases such as Alzheimer's disease (AD). To minimize subjectivity and inter-rater variability, an automated, accurate, precise, and reliable segmentation technique for the hippocampus (HC) and surrounding cerebro-spinal fluid (CSF) filled spaces-such as the temporal horns of the lateral ventricles-is essential. We trained and evaluated three automated methods for the segmentation of both HC and CSF (Multi-Atlas Label Fusion (MALF), Nonlinear Patch-Based Segmentation (NLPB), and a Convolutional Neural Network (CNN)). We then evaluated these methods, including the widely used FreeSurfer technique, using baseline T1w MRIs of 1641 participants from the AD Neuroimaging Initiative study with various degree of atrophy associated with their cognitive status on the spectrum from cognitively healthy to clinically probable AD. Our gold standard consisted in manual segmentation of HC and CSF from 80 cognitively healthy individuals. We calculated HC volumes and HVR and compared all methods in terms of segmentation reliability, similarity across methods, sensitivity in detecting between-group differences and associations with age, scores of the learning subtest of the Rey Auditory Verbal Learning Test (RAVLT) and the Alzheimer's Disease Assessment Scale 13 (ADAS13) scores. Cross validation demonstrated that the CNN method yielded more accurate HC and CSF segmentations when compared to MALF and NLPB, demonstrating higher volumetric overlap (Dice Kappa = 0.94) and correlation (rho = 0.99) with the manual labels. It was also the most reliable method in clinical data application, showing minimal failures. Our comparisons yielded high correlations between FreeSurfer, CNN and NLPB volumetric values. HVR yielded higher control:AD effect sizes than HC volumes among all segmentation methods, reinforcing the significance of HVR in clinical distinction. The positive association with age was significantly stronger for HVR compared to HC volumes on all methods except FreeSurfer. Memory associations with HC volumes or HVR were only significant for individuals with mild cognitive impairment. Finally, the HC volumes and HVR showed comparable negative associations with ADAS13, particularly in the mild cognitive impairment cohort. This study provides an evaluation of automated segmentation methods centered to estimate HVR, emphasizing the superior performance of a CNN-based algorithm. The findings underscore the pivotal role of accurate segmentation in HVR calculations for precise clinical applications, contributing valuable insights into medial temporal lobe atrophy in neurodegenerative disorders, especially AD.

Osteoporosis (OP), a systemic skeletal disease characterized by compromised bone strength and elevated fracture susceptibility, represents a growing global health challenge that necessitates early detection and accurate risk stratification. With the exponential growth of multidimensional biomedical data in OP research, feature selection has become an indispensable machine learning paradigm that improves model generalizability. At the same time, it preserves clinical interpretability and enhances predictive accuracy. This perspective article systematically reviews the transformative role of feature selection methodologies across three critical domains of OP investigation: 1) multi-omics biomarker identification, 2) diagnostic pattern recognition, and 3) fracture risk prognostication. In biomarker discovery, advanced feature selection algorithms systematically refine high-dimensional multi-omics datasets (genomic, proteomic, metabolomic) to isolate key molecular signatures correlated with bone mineral density (BMD) trajectories and microarchitectural deterioration. For clinical diagnostics, these techniques enable efficient extraction of discriminative pattern from multimodal imaging data, including dual-energy X-ray absorptiometry (DXA), quantitative computed tomography (CT), and emerging dental radiographic biomarkers. In prognostic modeling, strategic variable selection optimizes prognostic accuracy by integrating demographic, biochemical, and biomechanical predictors while migrating overfitting in heterogeneous patient cohorts. Current challenges include heterogeneity in dataset quality and dimensionality, translational gaps between algorithmic outputs and clinical decision parameters, and limited reproducibility across diverse populations. Future directions should prioritize the development of adaptive feature selection frameworks capable of dynamic multi-omics data integration, coupled with hybrid intelligence systems that synergize machine-derived biomarkers with clinician expertise. Addressing these challenges requires coordinated interdisciplinary efforts to establish standardized validation protocols and create clinician-friendly decision support interfaces, ultimately bridging the gap between computational OP research and personalized patient care.

To establish a comprehensive reference range for OxNNet-derived first-trimester placental volume (FTPV), based on values observed in healthy pregnancies. Data were obtained from the First Trimester Placental Ultrasound Study, an observational cohort study in which three-dimensional placental ultrasound imaging was performed between 11 + 2 and 14 + 1 weeks' gestation, alongside otherwise routine care. A subgroup of singleton pregnancies resulting in term live birth, without neonatal unit admission or major chromosomal or structural abnormality, were included. Exclusion criteria were fetal growth restriction, maternal diabetes mellitus, hypertensive disorders of pregnancy or other maternal medical conditions (e.g. chronic hypertension, antiphospholipid syndrome, systemic lupus erythematosus). Placental images were processed using the OxNNet toolkit, a software solution based on a fully convolutional neural network, for automated placental segmentation and volume calculation. Quantile regression and the lambda-mu-sigma (LMS) method were applied to model the distribution of FTPV, using both crown-rump length (CRL) and gestational age as predictors. Model fit was assessed using the Akaike information criterion (AIC), and centile curves were constructed for visual inspection. The cohort comprised 2547 cases. The distribution of FTPV across gestational ages was positively skewed, with variation in the distribution at different gestational timepoints. In model comparisons, the LMS method yielded lower AIC values compared with quantile regression models. For predicting FTPV from CRL, the LMS model with the Sinh-Arcsinh distribution achieved the best performance, with the lowest AIC value. For gestational-age-based prediction, the LMS model with the Box-Cox Cole and Green original distribution achieved the lowest AIC value. The LMS models were selected to construct centile charts for FTPV based on both CRL and gestational age. Evaluation of the centile charts revealed strong agreement between predicted and observed centiles, with minimal deviations. Both models demonstrated excellent calibration, and the Z-scores derived using each of the models confirmed normal distribution. This study established reference ranges for FTPV based on both CRL and gestational age in healthy pregnancies. The LMS method provided the best model fit, demonstrating excellent calibration and minimal deviations between predicted and observed centiles. These findings should facilitate the exploration of FTPV as a potential biomarker for adverse pregnancy outcome and provide a foundation for future research into its clinical applications. © 2025 The Author(s). Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.

The lumbar spine and pelvis are critical stress-bearing structures of the human body, and their rapid and accurate segmentation plays a vital role in clinical diagnosis and intervention. However, conventional CT imaging poses significant challenges due to the low contrast of sacral and bilateral hip tissues and the complex and highly similar intervertebral space structures within the lumbar spine. To address these challenges, we propose a general-purpose segmentation network that integrates a cross-scale feature fusion strategy with a linear self-attention mechanism. The proposed network effectively extracts multi-scale features and fuses them along the channel dimension, enabling both structural and boundary information of lumbar and pelvic regions to be captured within the encoder-decoder architecture.Furthermore, we introduce a linear mapping strategy to approximate the traditional attention matrix with a low-rank representation, allowing the linear attention mechanism to significantly reduce computational complexity while maintaining segmentation accuracy for vertebrae and pelvic bones. Comparative and ablation experiments conducted on the CTSpine1K and CTPelvic1K datasets demonstrate that our method achieves improvements of 1.5% in Dice Similarity Coefficient (DSC) and 2.6% in Hausdorff Distance (HD) over state-of-the-art models, validating the effectiveness of our approach in enhancing boundary segmentation quality and segmentation accuracy in homogeneous anatomical regions.

The objective of this study was to develop a deep learning model for automated pituitary adenoma segmentation in MRI scans for stereotactic radiosurgery planning and to assess its accuracy and efficiency in clinical settings. An nnU-Net-based model was trained on MRI scans with expert segmentations of 582 patients treated with Leksell Gamma Knife over the course of 12 years. The accuracy of the model was evaluated by a human expert on a separate dataset of 146 previously unseen patients. The primary outcome was the comparison of expert ratings between the predicted segmentations and a control group consisting of original manual segmentations. Secondary outcomes were the influence of tumor volume, previous surgery, previous stereotactic radiosurgery (SRS), and endocrinological status on expert ratings, performance in a subgroup of nonfunctioning macroadenomas (measuring 1000-4000 mm3) without previous surgery and/or radiosurgery, and influence of using additional MRI modalities as model input and time cost reduction. The model achieved Dice similarity coefficients of 82.3%, 63.9%, and 79.6% for tumor, normal gland, and optic nerve, respectively. A human expert rated 20.6% of the segmentations as applicable in treatment planning without any modifications, 52.7% as applicable with minor manual modifications, and 26.7% as inapplicable. The ratings for predicted segmentations were lower than for the control group of original segmentations (p < 0.001). Larger tumor volume, history of a previous radiosurgery, and nonfunctioning pituitary adenoma were associated with better expert ratings (p = 0.005, p = 0.007, and p < 0.001, respectively). In the subgroup without previous surgery, although expert ratings were more favorable, the association did not reach statistical significance (p = 0.074). In the subgroup of noncomplex cases (n = 9), 55.6% of the segmentations were rated as applicable without any manual modifications and no segmentations were rated as inapplicable. Manually improving inaccurate segmentations instead of creating them from scratch led to 53.6% reduction of the time cost (p < 0.001). The results were applicable for treatment planning with either no or minor manual modifications, demonstrating a significant increase in the efficiency of the planning process. The predicted segmentations can be loaded into the planning software used in clinical practice for treatment planning. The authors discuss some considerations of the clinical utility of the automated segmentation models, as well as their integration within established clinical workflows, and outline directions for future research.

The primary goal in neuro-oncology is the maximally safe resection of high-grade glioma (HGG). A more extensive resection improves both overall and disease-free survival, while a complication-free surgery enables better tolerance to adjuvant therapies such as chemotherapy and radiotherapy. Techniques such as 5-aminolevulinic acid (5-ALA) fluorescence and intraoperative ultrasound (ioUS) are valuable for safe resection and cost-effective. However, the benefits of combining these techniques remain undocumented. The aim of this study was to investigate outcomes when combining 5-ALA and ioUS. From January 2019 to January 2024, 72 patients (mean age 62.2 years, 62.5% male) underwent HGG resection at a single hospital. Tumor histology included glioblastoma (90.3%), grade IV astrocytoma (4.1%), grade III astrocytoma (2.8%), and grade III oligodendroglioma (2.8%). Tumor resection was performed under natural light, followed by using 5-ALA and ioUS to detect residual tumor. Biopsies from the surgical bed were analyzed for tumor presence and categorized based on 5-ALA and ioUS results. Results of 5-ALA and ioUS were classified into positive, weak/doubtful, or negative. Histological findings of the biopsies were categorized into solid tumor, infiltration, or no tumor. Sensitivity, specificity, and predictive values for both techniques, separately and combined, were calculated. A machine learning algorithm (HGGPredictor) was developed to predict tumor presence in biopsies. The overall sensitivities of 5-ALA and ioUS were 84.9% and 76%, with specificities of 57.8% and 84.5%, respectively. The combination of both methods in a positive/positive scenario yielded the highest performance, achieving a sensitivity of 91% and specificity of 86%. The positive/doubtful combination followed, with sensitivity of 67.9% and specificity of 95.2%. Area under the curve analysis indicated superior performance when both techniques were combined, in comparison to each method used individually. Additionally, the HGGPredictor tool effectively estimated the quantity of tumor cells in surgical margins. Combining 5-ALA and ioUS enhanced diagnostic accuracy for HGG resection, suggesting a new surgical standard. An intraoperative predictive algorithm could further automate decision-making.