Computed Tomography serves as an indispensable tool in clinical workflows, providing non-invasive visualization of internal anatomical structures. Existing CT reconstruction works are limited to small-capacity model architecture, inflexible volume representation, and small-scale training data. In this paper, we present X-GRM (X-ray Gaussian Reconstruction Model), a large feedforward model for reconstructing 3D CT from sparse-view 2D X-ray projections. X-GRM employs a scalable transformer-based architecture to encode an arbitrary number of sparse X-ray inputs, where tokens from different views are integrated efficiently. Then, tokens are decoded into a new volume representation, named Voxel-based Gaussian Splatting (VoxGS), which enables efficient CT volume extraction and differentiable X-ray rendering. To support the training of X-GRM, we collect ReconX-15K, a large-scale CT reconstruction dataset containing around 15,000 CT/X-ray pairs across diverse organs, including the chest, abdomen, pelvis, and tooth etc. This combination of a high-capacity model, flexible volume representation, and large-scale training data empowers our model to produce high-quality reconstructions from various testing inputs, including in-domain and out-domain X-ray projections. Project Page: https://github.com/CUHK-AIM-Group/X-GRM.

[¹⁸F]FDG PET/CT scan combined with [¹⁸F]PSMA-1007 PET/CT scan is commonly conducted for detecting bone metastases in prostate cancer (PCa). However, it is expensive and may expose patients to more radiation hazards. This study explores deep learning (DL) techniques to synthesize [¹⁸F]PSMA-1007 PET bone images from CT bone images for the early detection of bone metastases in PCa, which may reduce additional PET/CT scans and relieve the burden on patients. We retrospectively collected paired whole-body (WB) [¹⁸F]PSMA-1007 PET/CT images from 152 patients with clinical and pathological diagnosis results, including 123 PCa and 29 cases of benign lesions. The average age of the patients was 67.48 ± 10.87 years, and the average lesion size was 8.76 ± 15.5 mm. The paired low-dose CT and PET images were preprocessed and segmented to construct the WB bone structure images. 152 subjects were randomly stratified into training, validation, and test groups in the number of 92:41:19. Two generative adversarial network (GAN) models-Pix2pix and Cycle GAN-were trained to synthesize [¹⁸F]PSMA-1007 PET bone images from paired CT bone images. The performance of two synthesis models was evaluated using quantitative metrics of mean absolute error (MAE), mean squared error (MSE), peak signal-to-noise ratio (PSNR), and structural similarity index metrics (SSIM), as well as the target-to-background ratio (TBR). The results of DL-based image synthesis indicated that the synthesis of [¹⁸F]PSMA-1007 PET bone images from low-dose CT bone images was highly feasible. The Pix2pix model performed better with an SSIM of 0.97, PSNR of 44.96, MSE of 0.80, and MAE of 0.10, respectively. The TBRs of bone metastasis lesions calculated on DL-synthesized PET bone images were highly correlated with those of real PET bone images (Pearson's r > 0.90) and had no significant differences (p < 0.05). It is feasible to generate synthetic [¹⁸F]PSMA-1007 PET bone images from CT bone images by using DL techniques with reasonable accuracy, which can provide information for early detection of PCa bone metastases.

To perform a systematic review on the impact of deep learning (DL)-based triage for reducing diagnostic delays and improving patient outcomes in peer-reviewed and pre-print publications. A search was conducted of primary research studies focused on DL-based worklist optimization for diagnostic imaging triage published on multiple databases from January 2018 until July 2024. Extracted data included study design, dataset characteristics, workflow metrics including report turnaround time and time-to-treatment, and patient outcome differences. Further analysis between clinical settings and integration modality was investigated using nonparametric statistics. Risk of bias was assessed with the risk of bias in non-randomized studies-of interventions (ROBINS-I) checklist. A total of 38 studies from 20 publications, involving 138,423 images, were analyzed. Workflow interventions concerned pulmonary embolism (n = 8), stroke (n = 3), intracranial hemorrhage (n = 12), and chest conditions (n = 15). Patients in the post DL-triage group had shorter median report turnaround times: a mean difference of 12.3 min (IQR: -25.7, -7.6) for pulmonary embolism, 20.5 min (IQR: -32.1, -9.3) for stroke, 4.3 min (IQR: -8.6, 1.3) for intracranial hemorrhage and 29.7 min (IQR: -2947.7, -18.3) for chest diseases. Sub-group analysis revealed that reductions varied per clinical environment and relative prevalence rates but were the highest when algorithms actively stratified and reordered the radiological worklist, with reductions of -43.7% in report turnaround time compared to -7.6% from widget-based systems (p < 0.01). DL-based triage systems had comparable report turnaround time improvements, especially in outpatient and high-prevalence settings, suggesting that AI-based triage holds promise in alleviating radiology workloads. Question Can DL-based triage address lengthening imaging report turnaround times and improve patient outcomes across distinct clinical environments? Findings DL-based triage improved report turnaround time across disease groups, with higher reductions reported in high-prevalence or lower acuity settings. Clinical relevance DL-based workflow prioritization is a reliable tool for reducing diagnostic imaging delay for time-sensitive disease across clinical settings. However, further research and reliable metrics are needed to provide specific recommendations with regards to false-negative examinations and multi-condition prioritization.

Medical image captioning is a challenging task that requires generating clinically accurate and semantically meaningful descriptions of radiology images. While recent vision-language models (VLMs) such as BLIP, BLIP2, Gemini and ViT-GPT2 show strong performance on natural image datasets, they often produce generic or imprecise captions when applied to specialized medical domains. In this project, we explore the effectiveness of fine-tuning the BLIP model on the ROCO dataset for improved radiology captioning. We compare the fine-tuned BLIP against its zero-shot version, BLIP-2 base, BLIP-2 Instruct and a ViT-GPT2 transformer baseline. Our results demonstrate that domain-specific fine-tuning on BLIP significantly improves performance across both quantitative and qualitative evaluation metrics. We also visualize decoder cross-attention maps to assess interpretability and conduct an ablation study to evaluate the contributions of encoder-only and decoder-only fine-tuning. Our findings highlight the importance of targeted adaptation for medical applications and suggest that decoder-only fine-tuning (encoder-frozen) offers a strong performance baseline with 5% lower training time than full fine-tuning, while full model fine-tuning still yields the best results overall.

Pathological myopia (PM) has emerged as a leading cause of global visual impairment, early detection and precise grading of PM are crucial for timely intervention. The atrophy, traction and neovascularisation (ATN) system is applied to define PM progression and stages with precision. This study focuses on constructing a comprehensive PM image dataset comprising both fundus and optical coherence tomography (OCT) images and developing a bimodal artificial intelligence (AI) classification model for ATN grading in PM. This single-centre retrospective cross-sectional study collected 2760 colour fundus photographs and matching OCT images of PM from January 2019 to November 2022 at Peking Union Medical College Hospital. Ophthalmology specialists labelled and inspected all paired images using the ATN grading system. The AI model used a ResNet-50 backbone and a multimodal multi-instance learning module to enhance interaction across instances from both modalities. Performance comparisons among single-modality fundus, OCT and bimodal AI models were conducted for ATN grading in PM. The bimodality model, dual-deep learning (DL), demonstrated superior accuracy in both detailed multiclassification and biclassification of PM, which aligns well with our observation from instance attention-weight activation maps. The area under the curve for severe PM using dual-DL was 0.9635 (95% CI 0.9380 to 0.9890), compared with 0.9359 (95% CI 0.9027 to 0.9691) for the solely OCT model and 0.9268 (95% CI 0.8915 to 0.9621) for the fundus model. Our novel bimodal AI multiclassification model for PM ATN staging proves accurate and beneficial for public health screening and prompt referral of PM patients.

A layman in health systems is a person who doesn't have any knowledge about health data i.e., X-ray, MRI, CT scan, and health examination reports, etc. The motivation behind the proposed invention is to help laymen to make medical images understandable. The health model is trained using a neural network approach that analyses user health examination data; predicts the type and level of the disease and advises precaution to the user. Cellular Automata (CA) technology has been integrated with the neural networks to segment the medical image. The CA analyzes the medical images pixel by pixel and generates a robust threshold value which helps to efficiently segment the image and identify accurate abnormal spots from the medical image. The proposed method has been trained and experimented using 10000+ medical images which are taken from various open datasets. Various text analysis measures i.e., BLEU, ROUGE, and WER are used in the research to validate the produced report. The BLEU and ROUGE calculate a similarity to decide how the generated text report is closer to the original report. The BLEU and ROUGE scores of the experimented images are approximately 0.62 and 0.90, claims that the produced report is very close to the original report. The WER score 0.14, claims that the generated report contains the most relevant words. The overall summary of the proposed research is that it provides a fruitful medical report with accurate disease and precautions to the laymen.

Neuropsychiatric disorders have complex pathological mechanism, pronounced clinical heterogeneity, and a prolonged preclinical phase, which presents a challenge for early diagnosis and development of precise intervention strategies. With the development of large-scale multimodal neuroimaging datasets and advancement of artificial intelligence (AI) algorithms, the integration of multimodal imaging with AI techniques has emerged as a pivotal avenue for early detection and tailoring individualized treatment for neuropsychiatric disorders. To support these advances, in this review, we outline multimodal neuroimaging techniques, AI methods, and strategies for multimodal data fusion. We highlight applications of multimodal AI based on neuroimaging data in precision medicine for neuropsychiatric disorders, discussing challenges in clinical adoption, their emerging solutions, and future directions.

Artificial intelligence (AI) has emerged as a transformative force in orthopedic surgery. Potentially encompassing pre-, intra-, and postoperative processes, it can process complex medical imaging, provide real-time surgical guidance, and analyze large datasets for outcome prediction and optimization. AI has shown improvements in surgical precision, efficiency, and patient outcomes across orthopedic subspecialties, and large language models and agentic AI systems are expanding AI utility beyond surgical applications into areas such as clinical documentation, patient education, and autonomous decision support. The successful implementation of AI in orthopedic surgery requires careful attention to validation, regulatory compliance, and healthcare system integration. As these technologies continue to advance, maintaining the balance between innovation and patient safety remains crucial, with the ultimate goal of achieving more personalized, efficient, and equitable healthcare delivery while preserving the essential role of human clinical judgment. This review examines the current landscape and future trajectory of AI applications in orthopedic surgery, highlighting both technological advances and their clinical impact. Studies have suggested that AI-assisted procedures achieve higher accuracy and better functional outcomes compared to conventional methods, while reducing operative times and complications. However, these technologies are designed to augment rather than replace clinical expertise, serving as sophisticated tools to enhance surgeons' capabilities and improve patient care.

This study aims to investigate the feasibility of a single general model to synthesize CT images across body sites, thorax, abdomen, and pelvis, to support treatment planning for MRI-only radiotherapy. A total of 157 patients who received MRI-guided radiation therapy in the thorax, abdomen, and pelvis on a 0.35T MRIdian Linac were included. A subset of 122 cases were used for model training and the remaining 35 cases were used for model validation. All patient datasets had semi-paired CT-simulation image and 0.35T MR image acquired using TrueFISP. A conditional generative adversarial network with a multi-planar method was used to generate synthetic CT images from 0.35T MR images. The effect of preprocessing methods (with and without bias field corrections) on the quality of synthetic CT was evaluated and found to be insignificant. The general models trained on all cases performed comparably to the site-specific models trained on individual body sites. For all models, the peak signal-to-noise ratios ranged from 31.7 to 34.9 and the structural index similarity measures ranged from 0.9547 to 0.9758. For the datasets with bias field corrections, the mean-absolute-errors in HU (general model versus site-specific model) were 49.7 ± 9.4 versus 49.5 ± 8.9, 48.7 ± 7.6 versus 43 ± 7.8 and 32.8 ± 5.5 versus 31.8 ± 5.3 for the thorax, abdomen, and pelvis, respectively. When comparing plans between synthetic CTs and ground truth CTs, the dosimetric difference was on average less than 0.5% (0.2 Gy) for target coverage and less than 2.1% (0.4 Gy) for organ-at-risk metrics for all body sites with either the general or specific models. Synthetic CT plans showed good agreement with mean gamma pass rates of >94% and >99% for 1%/1 mm and 2%/2 mm, respectively. This study has demonstrated the feasibility of using a general model for multiple body sites and the potential of using synthetic CT to support an MRI-guided radiotherapy workflow.

Pancreatic Cancer (PC) is a highly aggressive tumor that is mainly diagnosed at later stages. Various imaging technologies, such as CT, MRI, and EUS, possess limitations in early PC diagnosis. Therefore, this review article explores the various innovative biomarkers for PC detection, such as DNA methylation, Noncoding RNAs, and proteomic biomarkers, and the role of AI in PC detection at early stages. Innovative biomarkers, such as DNA methylation genes, show higher specificity and sensitivity in PC diagnosis. Additionally, various non-coding RNAs, such as long non-coding RNAs (lncRNAs) and microRNAs, show high diagnostic accuracy and serve as diagnostic and prognostic biomarkers. Additionally, proteomic biomarkers retain higher diagnostic accuracy in different body fluids. Apart from this, the utilization of AI showed that AI surpassed the radiologist's diagnostic performance in PC detection. The combination of AI and advanced biomarkers can revolutionize early PC detection. However, large-scale, prospective studies are needed to validate its clinical utility. Further. standardization of biomarker panels and AI algorithms is a vital step toward their reliable applications in early PC detection, ultimately improving patient outcomes.