Since its development, virtual monoenergetic imaging (VMI) derived from dual-energy computed tomography (DECT) has been shown to be valuable in many clinical applications. However, DECT-based VMI showed increased noise at low keV levels. In this study, we proposed an unsupervised learning method to generate VMI from DECT. This means that we don't require training and labeled (i.e. high-quality VMI) data. Specifically, DECT images were fed into a deep learning (DL) based model expected to output VMI. Based on the theory that VMI obtained from image space data is a linear combination of DECT images, we used the model output (i.e. the predicted VMI) to recalculate DECT images. By minimizing the difference between the measured and recalculated DECT images, the DL-based model can be constrained itself to generate VMI from DECT images. We investigate whether the proposed DL-based method has the ability to improve the quality of VMIs. The experimental results obtained from patient data showed that the DL-based VMIs had better image quality than the conventional DECT-based VMIs. Moreover, the CT number differences between the DECT-based and DL-based VMIs were distributed within <math xmlns="http://www.w3.org/1998/Math/MathML"><mo>±</mo></math> 10 HU for bone and <math xmlns="http://www.w3.org/1998/Math/MathML"><mo>±</mo></math> 5 HU for brain, fat, and muscle. Except for bone, no statistically significant difference in CT number measurements was found between the DECT-based and DL-based VMIs (p > 0.01). Our preliminary results show that DL has the potential to unsupervisedly generate high-quality VMIs directly from DECT.

A diagnostic medical foundation model (MedFM) is an artificial intelligence (AI) system engineered to accurately determine diagnoses across various medical imaging modalities and specialties. To train MedFM, we created the PubMed Central Medical Images Dataset (PMCMID), the largest annotated medical image dataset to date, comprising 16,126,659 images from 3,021,780 medical publications. Using AI- and ontology-based methods, we identified 4,482,237 medical images (e.g., clinical photos, X-rays, ultrasounds) and generated comprehensive annotations. To optimize MedFMs performance and assess biases, 13,266 images were manually annotated to establish a multimodal benchmark. MedFM achieved physician-level performance in diagnosis tasks spanning radiology, dermatology, and infectious diseases without requiring specific training. Additionally, we developed the Image2Paper app, allowing clinicians to upload medical images and retrieve relevant literature. The correct diagnoses appeared within the top ten results in 88.4% and at least one relevant differential diagnosis in 93.0%. MedFM and PMCMID were made publicly available. FundingResearch reported here was partially supported by the National Cancer Institute (NCI) (R01 CA260271), the Saudi Company for Artificial Intelligence (SCAI) Authority, and the German Federal Ministry for Economic Affairs and Climate Action (BMWK) under the project DAKI-FWS (01MK21009E). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

This study aimed to evaluate the diagnostic performance of an artificial intelligence (AI)-based platform (Diagnocat) in detecting periapical radiolucencies (PARLs) in cone-beam computed tomography (CBCT) scans of molars. Specifically, we assessed Diagnocat's performance in detecting PARLs in non-root-filled molars and compared its diagnostic performance between preoperative and postoperative scans. This retrospective study analyzed preoperative and postoperative CBCT scans of 134 molars (327 roots). PARLs detected by Diagnocat were compared with assessments independently performed by two experienced endodontists, serving as the reference standard. Diagnostic performance was assessed at both tooth and root levels using sensitivity, specificity, accuracy, positive predictive value (PPV), negative predictive value (NPV), F1 score, and the area under the receiver operating characteristic curve (AUC-ROC). In preoperative scans of non-root-filled molars, Diagnocat demonstrated high sensitivity (teeth: 93.9%, roots: 86.2%), moderate specificity (teeth: 65.2%, roots: 79.9%), accuracy (teeth: 79.1%, roots: 82.6%), PPV (teeth: 71.8%, roots: 75.8%), NPV (teeth: 91.8%, roots: 88.8%), and F1 score (teeth: 81.3%, roots: 80.7%) for PARL detection. The AUC was 0.76 at the tooth level and 0.79 at the root level. Postoperative scans showed significantly lower PPV (teeth: 54.2%; roots: 46.9%) and F1 scores (teeth: 67.2%; roots: 59.2%). Diagnocat shows promise in detecting PARLs in CBCT scans of non-root-filled molars, demonstrating high sensitivity but moderate specificity, highlighting the need for human oversight to prevent overdiagnosis. However, diagnostic performance declined significantly in postoperative scans of root-filled molars. Further research is needed to optimize the platform's performance and support its integration into clinical practice. AI-based platforms such as Diagnocat can assist clinicians in detecting PARLs in CBCT scans, enhancing diagnostic efficiency and supporting decision-making. However, human expertise remains essential to minimize the risk of overdiagnosis and avoid unnecessary treatment.

To develop and validate an automated diabetic macular oedema (DME) classification system based on the images from different three-dimensional optical coherence tomography (3-D OCT) devices. A multicentre, platform-based development study using retrospective and cross-sectional data. Data were subjected to a two-level grading system by trained graders and a retina specialist, and categorised into three types: no DME, non-centre-involved DME and centre-involved DME (CI-DME). The 3-D convolutional neural networks algorithm was used for DME classification system development. The deep learning (DL) performance was compared with the diabetic retinopathy experts. Data were collected from Joint Shantou International Eye Center of Shantou University and the Chinese University of Hong Kong, Chaozhou People's Hospital and The Second Affiliated Hospital of Shantou University Medical College from January 2010 to December 2023. 7790 volumes of 7146 eyes from 4254 patients were annotated, of which 6281 images were used as the development set and 1509 images were used as the external validation set, split based on the centres. Accuracy, F1-score, sensitivity, specificity, area under receiver operating characteristic curve (AUROC) and Cohen's kappa were calculated to evaluate the performance of the DL algorithm. In classifying DME with non-DME, our model achieved an AUROCs of 0.990 (95% CI 0.983 to 0.996) and 0.916 (95% CI 0.902 to 0.930) for hold-out testing dataset and external validation dataset, respectively. To distinguish CI-DME from non-centre-involved-DME, our model achieved AUROCs of 0.859 (95% CI 0.812 to 0.906) and 0.881 (95% CI 0.859 to 0.902), respectively. In addition, our system showed comparable performance (Cohen's κ: 0.85 and 0.75) to the retina experts (Cohen's κ: 0.58-0.92 and 0.70-0.71). Our DL system achieved high accuracy in multiclassification tasks on DME classification with 3-D OCT images, which can be applied to population-based DME screening.

With the advancement of modern medicine and the development of technologies such as MRI, CT, and cellular analysis, it has become increasingly critical for clinicians to accurately interpret various diagnostic images. However, modern medical education often faces challenges due to limited access to high-quality teaching materials, stemming from privacy concerns and a shortage of educational resources (Balogh et al., 2015). In this context, image data generated by machine learning models, particularly generative models, presents a promising solution. These models can create diverse and comparable imaging datasets without compromising patient privacy, thereby supporting modern medical education. In this study, we explore the use of convolutional neural networks (CNNs) and CycleGAN (Zhu et al., 2017) for generating synthetic medical images. The source code is available at https://github.com/mliuby/COMP4211-Project.

Clinical decision-making routinely demands reasoning over heterogeneous data, yet existing multimodal language models (MLLMs) remain largely vision-centric and fail to generalize across clinical specialties. To bridge this gap, we introduce QoQ-Med-7B/32B, the first open generalist clinical foundation model that jointly reasons across medical images, time-series signals, and text reports. QoQ-Med is trained with Domain-aware Relative Policy Optimization (DRPO), a novel reinforcement-learning objective that hierarchically scales normalized rewards according to domain rarity and modality difficulty, mitigating performance imbalance caused by skewed clinical data distributions. Trained on 2.61 million instruction tuning pairs spanning 9 clinical domains, we show that DRPO training boosts diagnostic performance by 43% in macro-F1 on average across all visual domains as compared to other critic-free training methods like GRPO. Furthermore, with QoQ-Med trained on intensive segmentation data, it is able to highlight salient regions related to the diagnosis, with an IoU 10x higher than open models while reaching the performance of OpenAI o4-mini. To foster reproducibility and downstream research, we release (i) the full model weights, (ii) the modular training pipeline, and (iii) all intermediate reasoning traces at https://github.com/DDVD233/QoQ_Med.

Medical imaging is critical for diagnostics, but clinical adoption of advanced AI-driven imaging faces challenges due to patient variability, image artifacts, and limited model generalization. While deep learning has transformed image analysis, 3D medical imaging still suffers from data scarcity and inconsistencies due to acquisition protocols, scanner differences, and patient motion. Traditional augmentation uses a single pipeline for all transformations, disregarding the unique traits of each augmentation and struggling with large data volumes. To address these challenges, we propose a Multi-encoder Augmentation-Aware Learning (MEAL) framework that leverages four distinct augmentation variants processed through dedicated encoders. Three fusion strategies such as concatenation (CC), fusion layer (FL), and adaptive controller block (BD) are integrated to build multi-encoder models that combine augmentation-specific features before decoding. MEAL-BD uniquely preserves augmentation-aware representations, enabling robust, protocol-invariant feature learning. As demonstrated in a Computed Tomography (CT)-to-T1-weighted Magnetic Resonance Imaging (MRI) translation study, MEAL-BD consistently achieved the best performance on both unseen- and predefined-test data. On both geometric transformations (like rotations and flips) and non-augmented inputs, MEAL-BD outperformed other competing methods, achieving higher mean peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM) scores. These results establish MEAL as a reliable framework for preserving structural fidelity and generalizing across clinically relevant variability. By reframing augmentation as a source of diverse, generalizable features, MEAL supports robust, protocol-invariant learning, advancing clinically reliable medical imaging solutions.

Physics-driven artificial intelligence (PD-AI) reconstruction methods have emerged as the state-of-the-art for accelerating MRI scans, enabling higher spatial and temporal resolutions. However, the high resolution of these scans generates massive data volumes, leading to challenges in transmission, storage, and real-time processing. This is particularly pronounced in functional MRI, where hundreds of volumetric acquisitions further exacerbate these demands. Edge computing with FPGAs presents a promising solution for enabling PD-AI reconstruction near the MRI sensors, reducing data transfer and storage bottlenecks. However, this requires optimization of PD-AI models for hardware efficiency through quantization and bypassing traditional FFT-based approaches, which can be a limitation due to their computational demands. In this work, we propose a novel PD-AI computational MRI approach optimized for FPGA-based edge computing devices, leveraging 8-bit complex data quantization and eliminating redundant FFT/IFFT operations. Our results show that this strategy improves computational efficiency while maintaining reconstruction quality comparable to conventional PD-AI methods, and outperforms standard clinical methods. Our approach presents an opportunity for high-resolution MRI reconstruction on resource-constrained devices, highlighting its potential for real-world deployment.

Physics-driven deep learning (PD-DL) models have proven to be a powerful approach for improved reconstruction of rapid MRI scans. In order to train these models in scenarios where fully-sampled reference data is unavailable, self-supervised learning has gained prominence. However, its application at high acceleration rates frequently introduces artifacts, compromising image fidelity. To mitigate this shortcoming, we propose a novel way to train PD-DL networks via carefully-designed perturbations. In particular, we enhance the k-space masking idea of conventional self-supervised learning with a novel consistency term that assesses the model's ability to accurately predict the added perturbations in a sparse domain, leading to more reliable and artifact-free reconstructions. The results obtained from the fastMRI knee and brain datasets show that the proposed training strategy effectively reduces aliasing artifacts and mitigates noise amplification at high acceleration rates, outperforming state-of-the-art self-supervised methods both visually and quantitatively.

Recent advances in deep learning-based medical image registration have shown that training deep neural networks~(DNNs) does not necessarily require medical images. Previous work showed that DNNs trained on randomly generated images with carefully designed noise and contrast properties can still generalize well to unseen medical data. Building on this insight, we propose using registration between random images as a proxy task for pretraining a foundation model for image registration. Empirical results show that our pretraining strategy improves registration accuracy, reduces the amount of domain-specific data needed to achieve competitive performance, and accelerates convergence during downstream training, thereby enhancing computational efficiency.