Medical Image Analysis for Tumor Segmentation Using U-Net Variants

Achieving accurate bladder wall and tumor segmentation from MRI is critical for diagnosing and treating bladder cancer. However, automated segmentation remains challenging due to factors such as comparable density distributions, intricate tumor morphologies, and unclear boundaries. Considering the attributes of bladder MRI images, we propose an efficient multiscale hierarchical hybrid attention with a transformer (MH2AFormer) for bladder cancer and wall segmentation. Specifically, a multiscale hybrid attention and transformer (MHAT) module in the encoder is designed to adaptively extract and aggregate multiscale hybrid feature representations from the input image. In the decoder stage, we devise a multiscale hybrid attention (MHA) module to generate high-quality segmentation results from multiscale hybrid features. Combining these modules enhances the feature representation and guides the model to focus on tumor and wall regions, which helps to solve bladder image segmentation challenges. Moreover, MHAT utilizes the Fast Fourier Transformer with a large kernel (e.g., 224 × 224) to model global feature relationships while reducing computational complexity in the encoding stage. The model performance was evaluated on two datasets. As a result, the model achieves relatively best results regarding the intersection over union (IoU) and dice similarity coefficient (DSC) on both datasets (Dataset A: IoU = 80.26%, DSC = 88.20%; Dataset B: IoU = 89.74%, DSC = 94.48%). These advantageous outcomes substantiate the practical utility of our approach, highlighting its potential to alleviate the workload of radiologists when applied in clinical settings.

Interlayer information fusion-based and dual-attention improved U-Net for ABVS image sequence intelligent tumor segmentation

Accurate segmentation of tumor-infected regions in retinal Optical Coherence Tomography (OCT) images is critical for early diagnosis and clinical decision-making. However, conventional deep learning and transformer-based models often struggle to delineate overlapping and inter-dependent pixel features, leading to reduced segmentation precision. This study proposes a novel Dependent Inter-Feature Segmentation Method (DIFSM) to improve the localization and segmentation of retinal tumor regions in OCT images. The proposed DIFSM framework integrates advanced image preprocessing, inter-feature dependency analysis, and a Vision Transformer (ViT) architecture to distinguish differentiable and non-differentiable features. Overlapping pixel regions are identified through inter-feature intensity and gradient analysis, which are indicative of tumor-affected areas. The Vision Transformer is trained using matched and unmatched inter-feature representations to enhance contextual learning and resolve feature ambiguities. Experiments were conducted on the OCTID dataset comprising high-resolution retinal OCT images, and performance was evaluated using Dice coefficient, Intersection over Union (IoU), precision, sensitivity, specificity, and Mean Square Matching Error (MSME). Comparative analysis was performed against state-of-the-art segmentation models. The proposed DIFSM model achieved a Dice coefficient of 96.2% and an IoU of 94.8%, demonstrating excellent spatial overlap with expert-validated tumor regions. Precision, sensitivity, and specificity reached 96.8%, 96.6%, and 96.7%, respectively, while MSME was reduced to 6.11%. Compared to existing methods, DIFSM improved segmentation accuracy by 14.39%, precision by 14.11%, and reduced MSME by 13.5%. The model consistently outperformed benchmark approaches in detecting macular hole and central serous retinopathy-associated tumor regions while maintaining robustness to noise and structural variability. The proposed DIFSM framework effectively addresses the limitations of existing OCT segmentation methods by explicitly modeling inter-feature dependencies and resolving overlapping pixel ambiguities using a Vision Transformer. The significant improvements in segmentation accuracy and error reduction highlight its potential as a reliable and clinically applicable tool for automated retinal tumor detection in ophthalmic imaging. DIFSM offers a promising direction for enhancing OCT-based diagnostic systems and supporting ophthalmologists in early disease identification and treatment planning.

This study aims to overcome challenges in lumbar spine imaging, particularly lumbar spinal stenosis, by developing an automated segmentation model using advanced techniques. Traditional manual measurement and lesion detection methods are limited by subjectivity and inefficiency. The objective is to create an accurate and automated segmentation model that identifies anatomical structures in lumbar spine magnetic resonance imaging scans. Leveraging a dataset of 539 lumbar spinal stenosis patients, the study utilizes the residual U-Net for semantic segmentation in sagittal and axial lumbar spine magnetic resonance images. The model, trained to recognize specific tissue categories, employs a geometry algorithm for anatomical structure quantification. Validation metrics, like Intersection over Union (IOU) and Dice coefficients, validate the residual U-Net's segmentation accuracy. A novel rotation matrix approach is introduced for detecting bulging discs, assessing dural sac compression, and measuring yellow ligament thickness. The residual U-Net achieves high precision in segmenting lumbar spine structures, with mean IOU values ranging from 0.82 to 0.93 across various tissue categories and views. The automated quantification system provides measurements for intervertebral disc dimensions, dural sac diameter, yellow ligament thickness, and disc hydration. Consistency between training and testing datasets assures the robustness of automated measurements. Automated lumbar spine segmentation with residual U-Net and deep learning exhibits high precision in identifying anatomical structures, facilitating efficient quantification in lumbar spinal stenosis cases. The introduction of a rotation matrix enhances lesion detection, promising improved diagnostic accuracy, and supporting treatment decisions for lumbar spinal stenosis patients.

Brain tumor segmentation is the most important way to separate tumor areas from healthy brain tissue in medical imaging. This is necessary for making an accurate diagnosis and planning treatment. But building strong deep learning models is often hard because there isn't much labeled medical data available, and strict privacy rules stop data from being shared in one place. Federated Learning (FL) helps keep patient data private by keeping it local, but its performance often drops when data from different hospitals have big differences in quality, imaging protocols, and distribution. Our research seeks to create a privacy-preserving federated learning framework that adeptly manages significant data heterogeneity while ensuring high segmentation accuracy across various institutions. We propose a new two-stage FL framework that allows multiple institutions to work together while keeping their privacy and effectively dealing with complicated non-IID data distributions. To start, we use a Federated Masked Autoencoder (MAE) for self-supervised pre-training. This lets the model learn strong anatomical features from unlabeled MRI scans. Second, the model is carefully fine-tuned using an Attention ResUNet3D architecture to get very accurate tumor segmentation. We use the SCAFFOLD optimization algorithm to keep training stable across all clients, even when the scanner varies from site to site, thereby directly addressing client drift. We also use strategic foreground-biased sampling and Test-Time Augmentation (TTA) techniques to greatly improve segmentation accuracy in difficult, uneven tumor sub-regions. We ran extensive experiments on the BraTS 2024 dataset in simulated federated settings with 10, 50, and 100 different clients. The Dice coefficients we got were 0.826, 0.824, and 0.818, which demonstrate strong performance. In the end, these strong results show that the suggested method works well on a larger scale and can be used in a clinical setting.

Automated 3D breast ultrasound (ABUS) has substantial potential in breast imaging. ABUS appears to be beneficial because of its outstanding reproducibility and reliability, especially for screening women with dense breasts. However, due to the high number of slices in 3D ABUS, it requires lengthy screening time for radiologists, and they may miss small and subtle lesions. In this work, we propose to use a 3D Mask R-CNN method to automatically detect the location of the tumor and simultaneously segment the tumor contour. The performance of the proposed algorithm was evaluated using 25 patients' data with ABUS image and ground truth contours. To further access the performance of the proposed method, we quantified the intersection over union (IoU), Dice similarity coefficient (DSC), and center of mass distance (CMD) between the ground truth and segmentation. The resultant IoU 96% ± 2%, DSC 84% ± 3%, and CMD 1.95 ± 0.89 mm respectively, which demonstrated the high accuracy of tumor detection and 3D volume segmentation of the proposed Mask R-CNN method. We have developed a novel deep learning-based method and demonstrated its capability of being used as a useful tool for computer-aided diagnosis and treatment.

Recent studies on Parkinson's Disease (PD) discovered potential markers in brain MR images for its diagnosis and staging.One of them is the neuromelanin, which can be found in regions such as the substantia nigra and the locus coeruleus, both located in the brainstem.The analysis of these regions is normally conducted manually by a specialized professional, which can be costly and timeconsuming.Fortunately, there have been advances in automatic segmentation approaches in several areas.In medical imaging, U-Nets are showing state-of-the-art performance, but we did not find many works using this architecture to extract brainstem structures.A likely cause for this is the scarcity of large databases with well annotated segmentation of these areas.Thus, this paper explores the use of a dataset of our research group composed of T1-weighted MR images to train U-Net models with different types of slices to automatically demarcate the brainstem (as a first stage).The analysis starts with a scripted segmentation of the target region using the Freesurfer package.Thereby, we trained 4 models and evaluated them using Dice Similarity Coefficient (DSC) and Intersection over Union (IoU).Three models were trained on single anatomical planes, and one on multi-plane slices.Also, we applied grid search, varying optimizers, numbers of filters in the first layer, and learning rates.The best model trained with the larger subset with the best performance was the axial (DSC: 91.76% and IoU: 86.95%), followed by sagittal (DSC: 90.61% and IoU: 84.11%), coronal (DSC: 89.38% and IoU: 68.76%), and all slices (DSC: 16.36% and IoU: 7.58%).The next steps of our research are to differentiate the midbrain out of the brainstem, considering manual segmentation as ground truth, and test some approaches on determining the threshold and evaluating separately the 2 hemispheres of the midbrain.

Dental disorders can lead to serious implications such as heart attack or strokes if not diagnosed and treated early. The diagnosis of these disorders differs from dentist to dentist due to differences in perception, poor x-ray quality because of noise, and different types of patients. As a result, there is an urgent need to create automated, AI-driven diagnostic solutions for dental disorders. Deep learning solutions have shown outstanding results in automated medical image analysis tasks. Our work proposes a U-Net with attention blocks to segment teeth from dental panoramic X-rays. The proposed Attention U-Net consists of four encoding and decoding blocks and achieved Dice Coefficient, IOU score, Specificity, and F1 Score of 0.9318, 0.8724, 0.9910, and 0.9379, respectively. The Attention U-Net achieves a 1.7% better Dice Coefficient score and 2.9% better IOU (Intersection Over Union) score than one of the best segmentation models, the U-Net. The segmented output can be used in computer-aided diagnostic (CAD) systems to detect various mouth disorders, helping the dentist diagnose the problem efficiently and accurately.

Accurate intracranial artery segmentation from digital subtraction angiography (DSA) is critical for neurovascular diagnosis and intervention planning. Vascular extraction, which combines preprocessing methods and deep learning models, yields a high level of results, but limited preprocessing results constrain the improvement of results. We propose a texture-based contrast enhancement preprocessing framework integrated with the nnU-Net model to improve vessel segmentation in time-sequential DSA images. The method generates a combined feature mask by fusing local contrast, local entropy, and brightness threshold maps, which is then used as input for deep learning–based segmentation. Segmentation performance was evaluated using the DIAS dataset with various standard quantitative metrics. The proposed preprocessing significantly improved segmentation across all metrics compared to both the baseline and contrast-limited adaptive histogram equalization (CLAHE). Using nnU-Net, the method achieved a Dice Similarity Coefficient (DICE) of 0.83 ± 0.20 and an Intersection over Union (IoU) of 0.72 ± 0.14, outperforming CLAHE (DICE 0.79 ± 0.41, IoU 0.70 ± 0.23) and the baseline (DICE 0.65 ± 0.15, IoU 0.47 ± 0.20). Most notably, vessel connectivity (VC) dropped by over 65% relative to unprocessed images, indicating marked improvements in VC and topological accuracy. This study demonstrates that combining texture-based preprocessing with nnU-Net delivers robust, noise-tolerant, and clinically interpretable segmentation of intracranial arteries from DSA.

Convolutional neural networks have achieved excellent results in automatic medical image segmentation. In this study, we proposed a novel three-dimensional (3D) multipath DenseNet for generating the accurate glioblastoma (GBM) tumor contour from four multimodal pre-operative MR images. We hypothesized that the multipath architecture could achieve more accurate segmentation than a singlepath architecture. Two hundred and fifty-eight GBM patients were included in this study. Each patient had four MR images (T1-weighted, contrast-enhanced T1-weighted, T2-weighted, and FLAIR) and the manually segmented tumor contour. We built a 3D multipath DenseNet that could be trained to achieve an end-to-end mapping from four MR images to the corresponding GBM tumor contour. A 3D singlepath DenseNet was also built for comparison. Both DenseNets were based on the encoder-decoder architecture. All four images were concatenated and fed into a single encoder path in the singlepath DenseNet, while each input image had its own encoder path in the multipath DenseNet. The patient cohort was randomly split into a training set of 180 patients, a validation set of 39 patients, and a testing set of 39 patients. Model performance was evaluated using the Dice similarity coefficient (DSC), average surface distance (ASD), and 95% Hausdorff distance (HD95% ). Wilcoxon signed-rank tests were conducted to assess statistical significances. The singlepath DenseNet achieved the DSC of 0.911±0.060, ASD of 1.3±0.7mm, and HD95% of 5.2±7.1mm, while the multipath DenseNet achieved the DSC of 0.922±0.041, ASD of 1.1±0.5mm, and HD95% of 3.9±3.3mm. The P-values of all Wilcoxon signed-rank tests were less than 0.05. Both DenseNets generated GBM tumor contours in good agreement with the manually segmented contours from multimodal MR images. The multipath DenseNet achieved more accurate tumor segmentation than the singlepath DenseNet. Here presented the 3D multipath DenseNet that demonstrated an improved accuracy over comparable algorithms in the clinical task of GBM tumor segmentation.

Background/Objectives: Accurate nuclear segmentation is a fundamental step in computational pathology, enabling reliable estimation of cellularity and nuclear morphology. However, segmentation models are typically evaluated under ideal imaging conditions, while real-world microscopy data are affected by staining variability, noise, and image degradation. This study aims to comparatively evaluate three representative convolutional architectures for nuclei segmentation, with emphasis on robustness and clinical relevance under perturbed imaging conditions. Methods: U-Net, Attention U-Net, and U-Net++ were trained and evaluated on the BBBC038 nuclei microscopy dataset using fixed train–validation–test splits. Robustness was assessed under three types of synthetic perturbations: Gaussian blur, additive noise, and color jitter. Segmentation performance was quantified using the Dice coefficient and Intersection-over-Union (IoU). Paired Wilcoxon signed-rank tests with Holm correction and Cliff’s delta were used for statistical comparison. In addition, clinically relevant nuclear descriptors—nuclear count, median nuclear area, area interquartile range (IQR), and nuclear density—were extracted from predicted masks, and descriptor stability was analyzed as relative deviation from clean conditions. Results: Under clean imaging conditions, Attention U-Net achieved the highest mean Dice score, while paired statistical analysis indicated that U-Net++ exhibited the most consistent performance across test samples. Under image perturbations, Attention U-Net demonstrated greater robustness to blur and noise, whereas U-Net++ showed superior stability under color variations. Descriptor-based analysis further indicated that U-Net++ preserved nuclear count and density most reliably under chromatic perturbations, while U-Net exhibited larger instability in nuclear count and density, particularly under noise. Conclusions: Architectural design choices strongly influence not only pixel-level segmentation accuracy but also the stability of clinically relevant nuclear morphology descriptors. Robustness evaluation under multiple perturbation types reveals important trade-offs between architectures that are not captured by clean-image benchmarks alone. These findings highlight the necessity of multi-level evaluation strategies combining overlap metrics, statistical testing, robustness analysis, and descriptor stability assessment for future benchmarking and clinically reliable deployment of nuclei segmentation systems.

ScLNet: A cornea with scleral lens OCT layers segmentation dataset and new multi-task model

Accurate lesion segmentation is critical for cancer diagnosis and treatment. Convolutional neural networks (CNNs) are widely used for medical image segmentation but struggle to capture long-range dependencies. Transformers mitigate this limitation but come with high computational costs. Mamba, a state-space model (SSM), efficiently models long-range dependencies but lacks precision in fine details. To address these challenges, this study aimed to develop a novel segmentation approach that combines the strengths of CNNs, Transformers, and Mamba, enhancing both global context understanding and local feature extraction in medical image segmentation. We propose AttmNet, a U-shaped network designed for medical image segmentation, which incorporates a novel structure called MAM (Multiscale-Convolution, Self-Attention, and Mamba). The MAM block integrates multi-layer convolution for multi-scale feature learning with an Att-Mamba component that combines self-attention and Mamba to effectively capture global context while preserving fine details. We evaluated AttmNet on four public datasets for breast, skin, and lung lesion segmentation. AttmNet outperformed state-of-the-art methods in terms of intersection over union (IoU) and Dice similarity coefficients. On the breast ultrasound (BUS) dataset, AttmNet achieved a 3.38% improvement in IoU and a 4.54% increase in Dice over the next best method. On the breast ultrasound images (BUSI) dataset, AttmNet's IoU and Dice coefficients were 1.17% and 3.21% higher than the closest competitor, respectively. In the PH2 Dermoscopy Image dataset, AttmNet surpassed the next best model by 0.25% in both IoU and Dice. On the larger coronavirus disease 2019 (COVID-19) Lung dataset, AttmNet maintained strong performance, achieving higher IoU and Dice scores than the next best models, SegMamba and TransUNet. AttmNet is a powerful and efficient tool for medical image segmentation, addressing the limitations of existing methods through its advanced design. The MAM block significantly enhances segmentation accuracy while maintaining computational efficiency, making AttmNet highly suitable for clinical applications. The code is available at https://github.com/hyb2840/AttmNet.

Segmentation of the brain tumor on MRI is key to the exact diagnosis and treatment planning of patients suffering from a brain tumor. The process is also quite time-consuming and varies depending on the radiologist. For this reason, robust and accurate models for automated segmentation of brain tumors are in high demand. Through the following project work, robust and accurate brain tumor segmentation models are developed by the implementation of ResUNET and U-Net architectures and augmentation the data through optimized techniques. The hybrid model uses residual connections of ResUNET for deep feature extraction and the encoder-decoder structure of U-Net for precise segmentation. Elastic deformation techniques, in addition to augmentation parameters such as rotation, zooming, and changes in brightness, were simulated to mimic realistic variations in the tumor. Experiments show that the model trained with augmentation significantly outperforms the model trained without augmentation in terms of Dice score and IoU, showing better accuracy and robustness in segmentation. The model was trained and tested on the BraTS 2020 MRI dataset, in terms of performance evaluated by the Dice coefficient and Intersection over Union (IoU). In this work, the benefit of the association with advanced architectures to targeted augmentation techniques is demonstrated towards brain tumor segmentation, and hopes for it to support the work of radiologists in clinical settings in accurate identification and delimitation of boundaries in tumor MRI scans.

Latest advances in artificial intelligence, particularly in object recognition and segmentation, provide unprecedented opportunities for precision agriculture. This work investigates the use of state-of-the-art AI models, namely Meta’s Segment Anything (SAM) and GroundingDino, for the task of grape cluster detection in vineyards. Three different methods aimed at enhancing the instance segmentation process are proposed: (i) SAM-Refine (SAM-R), which refines a previously proposed depth-based clustering approach, referred to as DepthSeg, using SAM; (ii) SAM-Segmentation (SAM-S), which integrates SAM with a pre-trained semantic segmentation model to improve cluster separation; and (iii) AutoSAM-Dino (ASD), which eliminates the need for manual labeling and transfer learning through the combined use of GroundingDino and SAM. Analysis is conducted on both the object counting and pixel-level segmentation accuracy against a manually labeled ground truth. Metrics such as mean Average Precision (mAP), Intersection over Union (IoU), and precision and recall are calculated to assess the system performance. Compared to the original DepthSeg algorithm, SAM-R slightly advances object counting (mAP: +0.5%) and excels in pixel-level segmentation (IoU: +17.0%). SAM-S, despite a mAP decrease, improves segmentation accuracy (IoU: +13.9%, Precision: +9.2%, Recall: +11.7%). Similarly, ASD, although with a lower mAP, shows significant accuracy enhancement (IoU: +7.8%, Precision: +4.2%, Recall: +4.9%). Additionally, from a labor effort point of view, instance segmentation techniques require much less time for training than manual labeling.