Brain tumors, known for their life-threatening implications, underscore the urgency of precise and interpretable early detection. Expertise remains essential for accurate identification through MRI scans due to the intricacies involved. However, the growing recognition of automated detection systems holds the potential to enhance accuracy and improve interpretability. By consistently providing easily comprehensible results, these automated solutions could boost the overall efficiency and effectiveness of brain tumor diagnosis, promising a transformative era in healthcare. This paper introduces a new hybrid model, BioTransX, which uses a bi-former encoder mechanism, a dynamic sparse attention-based transformer, in conjunction with ensemble convolutional networks. Recognizing the importance of better contrast and data quality, we applied Contrast-Limited Adaptive Histogram Equalization (CLAHE) during the initial data processing stage. Additionally, to address the crucial aspect of model interpretability, we integrated Grad-CAM and Gradient Attention Rollout, which elucidate decisions by highlighting influential regions within medical images. Our hybrid deep learning model was primarily evaluated on the Kaggle MRI dataset for multi-class brain tumor classification, achieving a mean accuracy and F1-score of 99.29%. To validate its generalizability and robustness, BioTransX was further tested on two additional benchmark datasets, BraTS and Figshare, where it consistently maintained high performance across key evaluation metrics. The transformer-based hybrid model demonstrated promising performance in explainable identification and offered notable advantages in computational efficiency and memory usage. These strengths differentiate BioTransX from existing models in the literature and make it ideal for real-world deployment in resource-constrained clinical infrastructures.

Automated whole-brain delineation (WBD) techniques often struggle to generalize across pre-clinical studies due to variations in animal models, magnetic resonance imaging (MRI) scanners, and tissue contrasts. We developed a 3D U-Net neural network for WBD pre-trained on organophosphate intoxication (OPI) rat brain MRI scans. We used transfer learning (TL) to adapt this OPI-pretrained network to other animal models: rat model of Alzheimer's disease (AD), mouse model of tetramethylenedisulfotetramine (TETS) intoxication, and titi monkey model of social bonding. We assessed an OPI-pretrained 3D U-Net across animal models under three conditions: (1) direct application to each dataset; (2) utilizing TL; and (3) training disease-specific U-Net models. For each condition, training dataset size (TDS) was optimized, and output WBDs were compared to manual segmentations for accuracy. The OPI-pretrained 3D U-Net (TDS = 100) achieved the best accuracy [median[min-max]] for the test OPI dataset with a Dice coefficient (DC) = [0.987 [0.977-0.992]] and Hausdorff distance (HD) = [0.86 [0.55-1.27]]mm. TL improved generalization across all models [AD (TDS = 40): DC = 0.987 [0.977-0.992] and HD = 0.72 [0.54-1.00]mm; TETS (TDS = 10): DC = 0.992 [0.984-0.993] and HD = 0.40 [0.31-0.50]mm; Monkey (TDS = 8): DC = 0.977 [0.968-0.979] and HD = 3.03 [2.19-3.91]mm], showing performance comparable to disease-specific networks. The OPI-pretrained 3D U-Net with TL achieved accuracy comparable to disease-specific networks with reduced training data (TDS ≤ 40 scans) across all models. Future work will focus on developing a multi-region delineation pipeline for pre-clinical MRI brain data, utilizing the proposed WBD as an initial step.

Digital subtraction angiography (DSA) is an essential imaging technique in interventional radiology, enabling detailed visualization of blood vessels by subtracting pre- and post-contrast images. However, reduced contrast, either accidental or intentional, can impair the clarity of vascular structures. This issue becomes particularly critical in patients with chronic kidney disease (CKD), where minimizing iodinated contrast is necessary to reduce the risk of contrast-induced nephropathy (CIN). This study explored the potential of using a generative deep-learning-model based contrast enhancement technique for DSA. A text-conditioned image-to-image model was developed using Stable Diffusion, augmented with ControlNet to reduce hallucinations and Low-Rank Adaptation for model fine-tuning. A total of 1207 DSA series were used for training and testing, with additional low-contrast images generated through data augmentation. The model was trained using tagged text labels and evaluated using metrics such as Root Mean Square (RMS) contrast, Michelson contrast, signal-to-noise ratio (SNR), and entropy. Evaluation results indicated significant improvements, with RMS contrast, Michelson contrast, and entropy respectively increased from 7.91 to 17.7, 0.875 to 0.992, and 3.60 to 5.60, reflecting enhanced detail. However, SNR decreased from 21.3 to 8.50, indicating increased noise. This study demonstrated the feasibility of deep learning-based contrast enhancement for DSA images and highlights the potential for generative deep-learning-model to improve angiographic imaging. Further refinements, particularly in artifact suppression and clinical validation, are necessary for practical implementation in medical settings.

The COVID-19 pandemic plays a significant roles in the global health, highlighting the imperative for effective management of post-recovery symptoms. Within this context, Ground Glass Opacity (GGO) in lung computed tomography CT scans emerges as a critical indicator for early intervention. Recently, most researchers have investigated initially a challenge to refine techniques for GGO segmentation. These approaches aim to scrutinize and juxtapose cutting-edge methods for analyzing lung CT images of patients recuperating from COVID-19. While many methods in this challenge utilize the nnU-Net architecture, its general approach has not concerned completely GGO areas such as marking infected areas, ground-glass opacity, irregular shapes and fuzzy boundaries. This research has investigated a specialized machine learning algorithm, advancing the nn-UNet framework to accurately segment GGO in lung CT scans of post-COVID-19 patients. We propose a novel approach for two-stage image segmentation methods based on nnU-Net 2D and 3D models including lung and shadow image segmentation, incorporating the attention mechanism. The combination models enhance automatic segmentation and models' accuracy when using different error function in the training process. Experimental results show that the proposed model's outcomes DSC score ranks fifth among the compared results. The proposed method has also the second-highest sensitivity value among the methods, which shows that this method has a higher true segmentation rate than most of the other methods. The proposed method has achieved a Hausdorff95 of 54.566, Surface dice of 0.7193, Sensitivity of 0.7528, and Specificity of 0.7749. As compared with the state-of-the-art methods, the proposed model in experimental results is improved much better than the current methods in term of segmentation of infected areas. The proposed model has been deployed in the case study of real-world problems with the combination of 2D and 3D models. It is demonstrated the capacity to comprehensively detect lung lesions correctly. Additionally, the boundary loss function has assisted in achieving more precise segmentation for low-resolution images. Initially segmenting lung area has reduced the volume of images requiring processing, while diminishing for training process.

Deep neural networks are increasingly being used to detect and diagnose medical conditions using medical imaging. Despite their utility, these models are highly vulnerable to adversarial attacks and distribution shifts, which can affect diagnostic reliability and undermine trust among healthcare professionals. In this study, we propose a robust training algorithm with data augmentation (RTDA) to mitigate these vulnerabilities in medical image classification. We benchmark classifier robustness against adversarial perturbations and natural variations of RTDA and six competing baseline techniques, including adversarial training and data augmentation approaches in isolation and combination, using experimental data sets with three different imaging technologies (mammograms, X-rays, and ultrasound). We demonstrate that RTDA achieves superior robustness against adversarial attacks and improved generalization performance in the presence of distribution shift in each image classification task while maintaining high clean accuracy.

Generative Adversarial Networks (GAN) have shown potential in expanding limited medical imaging datasets. This study explores how different ratios of GAN-generated and real brain tumor MRI images impact the performance of a CNN in classifying healthy vs. tumorous scans. A DCGAN was used to create synthetic images which were mixed with real ones at various ratios to train a custom CNN. The CNN was then evaluated on a separate real-world test set. Our results indicate that the model maintains high sensitivity and precision in tumor classification, even when trained predominantly on synthetic data. When only a small portion of GAN data was added, such as 900 real images and 100 GAN images, the model achieved excellent performance, with test accuracy reaching 95.2%, and precision, recall, and F1-score all exceeding 95%. However, as the proportion of GAN images increased further, performance gradually declined. This study suggests that while GANs are useful for augmenting limited datasets especially when real data is scarce, too much synthetic data can introduce artifacts that affect the model's ability to generalize to real world cases.

Deep learning has demonstrated remarkable success in medical image segmentation and computer-aided diagnosis. In particular, numerous advanced methods have achieved state-of-the-art performance in brain tumor segmentation from MRI scans. While recent studies in other medical imaging domains have revealed that integrating textual reports with visual data can enhance segmentation accuracy, the field of brain tumor analysis lacks a comprehensive dataset that combines radiological images with corresponding textual annotations. This limitation has hindered the exploration of multimodal approaches that leverage both imaging and textual data. To bridge this critical gap, we introduce the TextBraTS dataset, the first publicly available volume-level multimodal dataset that contains paired MRI volumes and rich textual annotations, derived from the widely adopted BraTS2020 benchmark. Building upon this novel dataset, we propose a novel baseline framework and sequential cross-attention method for text-guided volumetric medical image segmentation. Through extensive experiments with various text-image fusion strategies and templated text formulations, our approach demonstrates significant improvements in brain tumor segmentation accuracy, offering valuable insights into effective multimodal integration techniques. Our dataset, implementation code, and pre-trained models are publicly available at https://github.com/Jupitern52/TextBraTS.

Pulmonary valve regurgitation after repaired Tetralogy of Fallot (TOF) or double-outlet right ventricle (DORV) causes hypertrophy and papillary muscle enlargement. Cardiac magnetic resonance imaging (CMR) can evaluate the right ventricular (RV) dilatation, but the effect of trabecular and papillary muscle (TPM) exclusion on RV volume for TOF or DORV reoperation decision is unclear. Twenty-three patients with repaired TOF or DORV, and 19 healthy controls aged ≥15, underwent CMR from 2012 to 2022. TPM volume is measured by artificial intelligence. Reoperation was considered when RV end-diastolic volume index (RVEDVI) >150 mL/m² or RV end-systolic volume index (RVESVI) >80 mL/m². RV volumes were higher in the disease group than controls (P α 0.001). RV mass and TPM volumes were higher in the disease group (P α 0.001). The reduction rate of RV volumes due to the exclusion of TPM volume was 6.3% (2.1-10.5), 11.7% (6.9-13.8), and 13.9% (9.5-19.4) in the control, volume load, and volume α pressure load groups, respectively. TPM/RV volumes were higher in the volume α pressure load group (control: 0.07 g/mL, volume: 0.14 g/mL, volume α pressure: 0.17 g/mL), and correlated with QRS duration (R α 0.77). In 3 patients in the volume α pressure, RV volume included TPM was indicated for reoperation, but when RV volume was reduced by TPM removal, reoperation was no indicated. RV volume measurements, including TPM in volume α pressure load, may help determine appropriate volume recommendations for reoperation.

Cone-beam computed tomography (CBCT) using only a few X-ray projection views enables faster scans with lower radiation dose, but the resulting severe under-sampling causes strong artifacts and poor spatial coverage. We address these challenges in a unified framework. First, we replace conventional UNet/ResNet encoders with TransUNet, a hybrid CNN-Transformer model. Convolutional layers capture local details, while self-attention layers enhance global context. We adapt TransUNet to CBCT by combining multi-scale features, querying view-specific features per 3D point, and adding a lightweight attenuation-prediction head. This yields Trans-CBCT, which surpasses prior baselines by 1.17 dB PSNR and 0.0163 SSIM on the LUNA16 dataset with six views. Second, we introduce a neighbor-aware Point Transformer to enforce volumetric coherence. This module uses 3D positional encoding and attention over k-nearest neighbors to improve spatial consistency. The resulting model, Trans$^2$-CBCT, provides an additional gain of 0.63 dB PSNR and 0.0117 SSIM. Experiments on LUNA16 and ToothFairy show consistent gains from six to ten views, validating the effectiveness of combining CNN-Transformer features with point-based geometry reasoning for sparse-view CBCT reconstruction.

Deep learning has demonstrated remarkable success in medical image segmentation and computer-aided diagnosis. In particular, numerous advanced methods have achieved state-of-the-art performance in brain tumor segmentation from MRI scans. While recent studies in other medical imaging domains have revealed that integrating textual reports with visual data can enhance segmentation accuracy, the field of brain tumor analysis lacks a comprehensive dataset that combines radiological images with corresponding textual annotations. This limitation has hindered the exploration of multimodal approaches that leverage both imaging and textual data. To bridge this critical gap, we introduce the TextBraTS dataset, the first publicly available volume-level multimodal dataset that contains paired MRI volumes and rich textual annotations, derived from the widely adopted BraTS2020 benchmark. Building upon this novel dataset, we propose a novel baseline framework and sequential cross-attention method for text-guided volumetric medical image segmentation. Through extensive experiments with various text-image fusion strategies and templated text formulations, our approach demonstrates significant improvements in brain tumor segmentation accuracy, offering valuable insights into effective multimodal integration techniques. Our dataset, implementation code, and pre-trained models are publicly available at https://github.com/Jupitern52/TextBraTS.