In recent years, deep learning has dominated medical image segmentation. Encoder-decoder architectures, such as U-Net, can be used in state-of-the-art models with powerful designs that are achieved by implementing skip connections that propagate local information from an encoder path to a decoder path to retrieve detailed spatial information lost by pooling operations. Despite their effectiveness for segmentation, these naïve skip connections still have some disadvantages. First, multi-scale skip connections tend to use unnecessary information and computational sources, where likable low-level encoder features are repeatedly used at multiple scales. Second, the contextual information of the low-level encoder feature is insufficient, leading to poor performance for pixel-wise recognition when concatenating with the corresponding high-level decoder feature. In this study, we propose a novel spatial-channel attention gate that addresses the limitations of plain skip connections. This can be easily integrated into an encoder-decoder network to effectively improve the performance of the image segmentation task. Comprehensive results reveal that our spatial-channel attention gate remarkably enhances the segmentation capability of the U-Net architecture with a minimal computational overhead added. The experimental results show that our proposed method outperforms the conventional deep networks in term of Dice score, which achieves 71.72%.

If left untreated, Alzheimer's disease (AD) is a leading cause of slowly progressive dementia. Therefore, it is critical to detect AD to prevent its progression. In this study, we propose a bidirectional progressive recurrent network with imputation (BiPro) that uses longitudinal data, including patient demographics and biomarkers of magnetic resonance imaging (MRI), to forecast clinical diagnoses and phenotypic measurements at multiple timepoints. To compensate for missing observations in the longitudinal data, we use an imputation module to inspect both temporal and multivariate relations associated with the mean and forward relations inherent in the time series data. To encode the imputed information, we define a modification of the long short-term memory (LSTM) cell by using a progressive module to compute the progression score of each biomarker between the given timepoint and the baseline through a negative exponential function. These features are used for the prediction task. The proposed system is an end-to-end deep recurrent network that can accomplish multiple tasks at the same time, including (1) imputing missing values, (2) forecasting phenotypic measurements, and (3) predicting the clinical status of a patient based on longitudinal data. We experimented on 1,335 participants from The Alzheimer's Disease Prediction of Longitudinal Evolution (TADPOLE) challenge cohort. The proposed method achieved a mean area under the receiver-operating characteristic curve (mAUC) of 78% for predicting the clinical status of patients, a mean absolute error (MAE) of 3.5ml for forecasting MRI biomarkers, and an MAE of 6.9ml for missing value imputation. The results confirm that our proposed model outperforms prevalent approaches, and can be used to minimize the progression of Alzheimer's disease.

Alzheimer's disease (AD) is a global neurodegenerative disorder that affects millions of individuals worldwide. Actual AD imaging datasets challenge the construction of reliable longitudinal models owing to imaging modality uncertainty. In addition, they are still unable to retain or obtain important information during disease progression from previous to followup time points. For example, the output values of current gates in recurrent models should be close to a specific value that indicates the model is uncertain about retaining or forgetting information. In this study, we propose a model which can extract and constrain each modality into a common representation space to capture intermodality interactions among different modalities associated with modality uncertainty to predict AD progression. In addition, we provide an auxiliary function to enhance the ability of recurrent gate robustly and effectively in controlling the flow of information over time using longitudinal data. We conducted comparative analysis on data from the Alzheimer's Disease Neuroimaging Initiative database. Our model outperformed other methods across all evaluation metrics. Therefore, the proposed model provides a promising solution for addressing modality uncertainty challenges in multimodal longitudinal AD progression prediction.

Accurate tissue segmentation is crucial for advancing healthcare, particularly in disease prediction and treatment planning. Precisely identifying abnormal tissue locations is a critical step for clinical analysis. While medical image segmentation increasingly utilizes multimodal and three-dimensional (3D) information to capture spatial relationships, current methods often struggle to effectively learn complementary information from multiple inputs, especially with complex 3D structures. In this study, we introduce AttCo, a novel multimodal semantic segmentation network built upon an attention-based co-learning fusion of deep feature representations. AttCo first employs multiple encoder branches to extract unimodal 3D representations from each imaging modality. These unimodal representations are then processed by a co-learning fusion module, which integrates both intra-modality (using SEAT) and inter-modality (using OSCAT) feature learning components. This dual approach ensures the capture of intricate interactions within each modality and across different modalities. Finally, these fused features pass through an up-sampling module to generate 3D segmented tumor maps. Our end-to-end deep network effectively addresses two key aspects: (i) extracting robust unimodal 3D representations and (ii) exploiting comprehensive inter- and intra-modality feature interactions. Experimental results demonstrate that AttCo significantly outperforms competing methods in terms of Dice score across various datasets. The source code can be found here https://github.com/duyphuongcri/AttCo.

Alzheimers disease progresses slowly and involves complex interaction between various biological factors. Longitudinal medical imaging data can capture this progression over time. However, longitudinal data frequently encounter issues such as missing data due to patient dropouts, irregular follow-up intervals, and varying lengths of observation periods. To address these issues, we designed a diffusion-based model for 3D longitudinal medical imaging generation using single magnetic resonance imaging (MRI). This involves the injection of a conditioning MRI and time-visit encoding to the model, enabling control in change between source and target images. The experimental results indicate that the proposed method generates higher-quality images compared to other competing methods.

Automated radiology report generation requires localized visual evidence and broader anatomical context. We propose ADAPT, an anatomy-guided framework that constructs compact visual prompts for large language model (LLM)-based report generation. Anatomy-Guided Token Pooling (AGTP) introduces anatomical segmentation masks as soft additive biases in cross-attention, steering learnable region queries toward relevant patches while retaining access to the complete patch sequence. A Dynamic Scale Predictor (DSP) estimates the bias strength for each examination, region, and attention head from mask-weighted visual features. The resulting tokens are projected into a frozen LLM to generate reports. Experiments on the Indiana University Chest X-Ray (IU-Xray) and Medical Information Mart for Intensive Care–Chest X-Ray (MIMIC-CXR) datasets show gains over an R2GenGPT baseline reproduced under the same training protocol, with all reported NLG and clinical label-agreement metrics improving on MIMIC-CXR and six of seven NLG metrics improving on IU-Xray. On MIMIC-CXR, ADAPT improves the Bilingual Evaluation Understudy (BLEU) scores BLEU-3 and BLEU-4 from 0.163 and 0.117 to 0.184 and 0.134, respectively, and CheXbert precision from 0.506 to 0.545. On IU-Xray, it achieves the highest Consensus-based Image Description Evaluation (CIDEr) score among the compared methods reporting that metric. Component ablations show that Region Token Pooling (RTP) accounts for much of the improvement over the baseline. On MIMIC-CXR, dynamic anatomical guidance yields the highest NLG point estimates among the tested fixed-scale alternatives, while learned-scale analysis reveals region- and input-dependent modulation. These findings support compact region-organized pooling with examination-conditioned anatomical guidance as an effective approach to visual-prompt construction.