We propose a compositional graph-based Machine Learning (ML) framework for Alzheimer's disease (AD) detection that constructs complex ML predictors from modular components. In our directed computational graph, datasets are represented as nodes [Formula: see text], and deep learning (DL) models are represented as directed edges [Formula: see text], allowing us to model complex image-processing pipelines [Formula: see text] as end-to-end DL predictors. Each directed path in the graph functions as a DL predictor, supporting both forward propagation for transforming data representations, as well as backpropagation for model finetuning, saliency map computation, and input data optimization. We demonstrate our model on Alzheimer's disease prediction, a complex problem that requires integrating multimodal data containing scans of different modalities and contrasts, genetic data and cognitive tests. We built a graph of 11 nodes (data) and 14 edges (ML models), where each model has been trained on handling a specific task (e.g. skull-stripping MRI scans, AD detection,image2image translation, ...). By using a modular and adaptive approach, our framework effectively integrates diverse data types, handles distribution shifts, and scales to arbitrary complexity, offering a practical tool that remains accurate even when modalities are missing for advancing Alzheimer's disease diagnosis and potentially other complex medical prediction tasks.

The rapid integration of artificial intelligence (AI) into healthcare has enhanced diagnostic accuracy; however, patient engagement and satisfaction remain significant challenges that hinder the widespread acceptance and effectiveness of AI-driven clinical tools. This study introduces CareAssist-GPT, a novel AI-assisted diagnostic model designed to improve both diagnostic accuracy and the patient experience through real-time, understandable, and empathetic communication. CareAssist-GPT combines high-resolution X-ray images, real-time physiological vital signs, and clinical notes within a unified predictive framework using deep learning. Feature extraction is performed using convolutional neural networks (CNNs), gated recurrent units (GRUs), and transformer-based NLP modules. Model performance was evaluated in terms of accuracy, precision, recall, specificity, and response time, alongside patient satisfaction through a structured user feedback survey. CareAssist-GPT achieved a diagnostic accuracy of 95.8%, improving by 2.4% over conventional models. It reported high precision (94.3%), recall (93.8%), and specificity (92.7%), with an AUC-ROC of 0.97. The system responded within 500 ms-23.1% faster than existing tools-and achieved a patient satisfaction score of 9.3 out of 10, demonstrating its real-time usability and communicative effectiveness. CareAssist-GPT significantly enhances the diagnostic process by improving accuracy and fostering patient trust through transparent, real-time explanations. These findings position it as a promising patient-centered AI solution capable of transforming healthcare delivery by bridging the gap between advanced diagnostics and human-centered communication.

Deep convolutional neural networks (CNNs) have seen significant growth in medical image classification applications due to their ability to automate feature extraction, leverage hierarchical learning, and deliver high classification accuracy. However, Deep CNNs require substantial computational power and memory, particularly for large datasets and complex architectures. Additionally, optimising the hyperparameters of deep CNNs, although critical for enhancing model performance, is challenging due to the high computational costs involved, making it difficult without access to high-performance computing resources. To address these limitations, this study presents a fast and efficient model that aims to achieve superior classification performance compared to popular Deep CNNs by developing lightweight CNNs combined with the Nonlinear Lévy chaotic moth flame optimiser (NLCMFO) for automatic hyperparameter optimisation. NLCMFO integrates the Lévy flight, chaotic parameters, and nonlinear control mechanisms to enhance the exploration capabilities of the Moth Flame Optimiser during the search phase while also leveraging the Lévy flight theorem to improve the exploitation phase. To assess the efficiency of the proposed model, empirical analyses were performed using a dataset of 2314 brain tumour detection images (1245 images of brain tumours and 1069 normal brain images). The evaluation results indicate that the CNN_NLCMFO outperformed a non-optimised CNN by 5% (92.40% accuracy) and surpassed established models such as DarkNet19 (96.41%), EfficientNetB0 (96.32%), Xception (96.41%), ResNet101 (92.15%), and InceptionResNetV2 (95.63%) by margins ranging from 1 to 5.25%. The findings demonstrate that the lightweight CNN combined with NLCMFO provides a computationally efficient yet highly accurate solution for medical image classification, addressing the challenges associated with traditional deep CNNs.

To develop an automated grading model for rectocele (RC) based on radiomics and evaluate its efficacy. This study retrospectively analyzed a total of 9,392 magnetic resonance imaging (MRI) images obtained from 222 patients who underwent dynamic magnetic resonance defecography (DMRD) over the period from August 2021 to June 2023. The focus was specifically on the defecation phase images of the DMRD, as this phase provides critical information for assessing RC. To develop and evaluate the model, the MRI images from all patients were randomly divided into two groups. 70% of the data were allocated to the training cohort to build the model, and the remaining 30% was reserved as a test cohort to evaluate its performance. First, the severity of RC was assessed using the RC MRI grading criteria by two independent radiologists. To extract and select radiomic features, two additional radiologists independently delineated the regions of interest (ROIs). These features were then dimensionality reduced to retain only the most relevant data for the analysis. The radiomics features were reduced in dimension, and a machine learning model was developed using a Support Vector Machine (SVM). Finally, receiver operating characteristic curve (ROC) and area under the curve (AUC) were used to evaluate the classification efficiency of the model. The AUC (macro/micro) of the model using defecation phase images was 0.794/0.824, and the overall accuracy was 0.754. The radiomics model built using the combination of DMRD defecation phase images is well suited for grading RC and helping clinicians diagnose and treat the disease.

Accurate preoperative assessment of lymph node metastasis (LNM) and overall survival (OS) status is essential for patients with locally advanced gastric cancer receiving neoadjuvant chemotherapy, providing timely guidance for clinical decision-making. However, current approaches to evaluate LNM and OS have limited accuracy. In this study, we used longitudinal CT images from 1,021 patients with locally advanced gastric cancer to develop and validate a multitask deep learning model, named co-attention tri-oriented spatial Mamba (CTSMamba), to simultaneously predict LNM and OS. CTSMamba was trained and validated on 398 patients, and the performance was further validated on 623 patients at two additional centers. Notably, CTSMamba exhibited significantly more robust performance than a clinical model in predicting LNM across all of the cohorts. Additionally, integrating CTSMamba survival scores with clinical predictors further improved personalized OS prediction. These results support the potential of CTSMamba to accurately predict LNM and OS from longitudinal images, potentially providing clinicians with a tool to inform individualized treatment approaches and optimized prognostic strategies. CTSMamba is a multitask deep learning model trained on longitudinal CT images of neoadjuvant chemotherapy-treated locally advanced gastric cancer that accurately predicts lymph node metastasis and overall survival to inform clinical decision-making. This article is part of a special series: Driving Cancer Discoveries with Computational Research, Data Science, and Machine Learning/AI.

Low-dose computed tomography (LDCT) imaging employed in lung cancer screening (LCS) programs is increasing in uptake worldwide. LCS programs herald a generational opportunity to simultaneously detect cancer and non-cancer-related early-stage lung disease. Yet these efforts are hampered by a shortage of radiologists to interpret scans at scale. Here, we present TANGERINE, a computationally frugal, open-source vision foundation model for volumetric LDCT analysis. Designed for broad accessibility and rapid adaptation, TANGERINE can be fine-tuned off the shelf for a wide range of disease-specific tasks with limited computational resources and training data. Relative to models trained from scratch, TANGERINE demonstrates fast convergence during fine-tuning, thereby requiring significantly fewer GPU hours, and displays strong label efficiency, achieving comparable or superior performance with a fraction of fine-tuning data. Pretrained using self-supervised learning on over 98,000 thoracic LDCTs, including the UK's largest LCS initiative to date and 27 public datasets, TANGERINE achieves state-of-the-art performance across 14 disease classification tasks, including lung cancer and multiple respiratory diseases, while generalising robustly across diverse clinical centres. By extending a masked autoencoder framework to 3D imaging, TANGERINE offers a scalable solution for LDCT analysis, departing from recent closed, resource-intensive models by combining architectural simplicity, public availability, and modest computational requirements. Its accessible, open-source lightweight design lays the foundation for rapid integration into next-generation medical imaging tools that could transform LCS initiatives, allowing them to pivot from a singular focus on lung cancer detection to comprehensive respiratory disease management in high-risk populations.

Vision transformers (ViTs) have rapidly gained prominence in medical imaging tasks such as disease classification, segmentation, and detection due to their superior accuracy compared to conventional deep learning models. However, due to their size and complex interactions via the self-attention mechanism, they are not well understood. In particular, it is unclear whether the representations produced by such models are semantically meaningful. In this paper, using a projected gradient-based algorithm, we show that their representations are not semantically meaningful and they are inherently vulnerable to small changes. Images with imperceptible differences can have very different representations; on the other hand, images that should belong to different semantic classes can have nearly identical representations. Such vulnerability can lead to unreliable classification results; for example, unnoticeable changes cause the classification accuracy to be reduced by over 60\%. %. To the best of our knowledge, this is the first work to systematically demonstrate this fundamental lack of semantic meaningfulness in ViT representations for medical image classification, revealing a critical challenge for their deployment in safety-critical systems.

We present our solution for the Multi-Source COVID-19 Detection Challenge, which classifies chest CT scans from four distinct medical centers. To address multi-source variability, we employ the Spatial-Slice Feature Learning (SSFL) framework with Kernel-Density-based Slice Sampling (KDS). Our preprocessing pipeline combines lung region extraction, quality control, and adaptive slice sampling to select eight representative slices per scan. We compare EfficientNet and Swin Transformer architectures on the validation set. The EfficientNet model achieves an F1-score of 94.68%, compared to the Swin Transformer's 93.34%. The results demonstrate the effectiveness of our KDS-based pipeline on multi-source data and highlight the importance of dataset balance in multi-institutional medical imaging evaluation.

The use of deep learning (DL) in medical image analysis has significantly improved the ability to predict lung cancer. In this study, we introduce a novel deep convolutional neural network (CNN) model, named ResNet+, which is based on the established ResNet framework. This model is specifically designed to improve the prediction of lung cancer and diseases using the images. To address the challenge of missing feature information that occurs during the downsampling process in CNNs, we integrate the ResNet-D module, a variant designed to enhance feature extraction capabilities by modifying the downsampling layers, into the traditional ResNet model. Furthermore, a convolutional attention module was incorporated into the bottleneck layers to enhance model generalization by allowing the network to focus on relevant regions of the input images. We evaluated the proposed model using five public datasets, comprising lung cancer (LC2500 $n$=3183, IQ-OTH/NCCD $n$=1336, and LCC $n$=25000 images) and lung disease (ChestXray $n$=5856, and COVIDx-CT $n$=425024 images). To address class imbalance, we used data augmentation techniques to artificially increase the representation of underrepresented classes in the training dataset. The experimental results show that ResNet+ model demonstrated remarkable accuracy/F1, reaching 98.14/98.14\% on the LC25000 dataset and 99.25/99.13\% on the IQ-OTH/NCCD dataset. Furthermore, the ResNet+ model saved computational cost compared to the original ResNet series in predicting lung cancer images. The proposed model outperformed the baseline models on publicly available datasets, achieving better performance metrics. Our codes are publicly available at https://github.com/AIPMLab/Graduation-2024/tree/main/Peng.

This study aimed to develop a pretreatment CT-based multichannel predictor integrating deep learning features encoded by Transformer models for preoperative diagnosis of major pathological response (MPR) in non-small cell lung cancer (NSCLC) patients receiving neoadjuvant immunochemotherapy. This multicenter diagnostic study retrospectively included 332 NSCLC patients from four centers. Pretreatment computed tomography images were preprocessed and segmented into region of interest cubes for radiomics modeling. These cubes were cropped into four groups of 2 dimensional image modules. GoogLeNet architecture was trained independently on each group within a multichannel framework, with gradient-weighted class activation mapping and SHapley Additive exPlanations value‌ for visualization. Deep learning features were carefully extracted and fused across the four image groups using the Transformer fusion model. After models training, model performance was evaluated via the area under the curve (AUC), sensitivity, specificity, F1 score, confusion matrices, calibration curves, decision curve analysis, integrated discrimination improvement, net reclassification improvement, and DeLong test. The dataset was allocated into training (n = 172, Center 1), internal validation (n = 44, Center 1), and external test (n = 116, Centers 2-4) cohorts. Four optimal deep learning models and the best Transformer fusion model were developed. In the external test cohort, traditional radiomics model exhibited an AUC of 0.736 [95% confidence interval (CI): 0.645-0.826]. The‌ optimal deep learning imaging ‌module‌ showed superior AUC of 0.855 (95% CI: 0.777-0.934). The fusion model named Transformer_GoogLeNet further improved classification accuracy (AUC = 0.924, 95% CI: 0.875-0.973). The new method of fusing multichannel deep learning with the Transformer Encoder can accurately diagnose whether NSCLC patients receiving neoadjuvant immunochemotherapy will achieve MPR. Our findings may support improved surgical planning and contribute to better treatment outcomes through more accurate preoperative assessment.