The increasing digitalization of multi-modal data in medicine and novel artificial intelligence (AI) algorithms opens up a large number of opportunities for predictive models. In particular, deep learning models show great performance in the medical field. A major limitation of such powerful but complex models originates from their 'black-box' nature. Recently, a variety of explainable AI (XAI) methods have been introduced to address this lack of transparency and trust in medical AI. However, the majority of such methods have solely been evaluated on single data modalities. Meanwhile, with the increasing number of XAI methods, integrative XAI frameworks and benchmarks are essential to compare their performance on different tasks. For that reason, we developed BenchXAI, a novel XAI benchmarking package supporting comprehensive evaluation of fifteen XAI methods, investigating their robustness, suitability, and limitations in biomedical data. We employed BenchXAI to validate these methods in three common biomedical tasks, namely clinical data, medical image and signal data, and biomolecular data. Our newly designed sample-wise normalization approach for post-hoc XAI methods enables the statistical evaluation and visualization of performance and robustness. We found that the XAI methods Integrated Gradients, DeepLift, DeepLiftShap, and GradientShap performed well over all three tasks, while methods like Deconvolution, Guided Backpropagation, and LRP-α1-β0 struggled for some tasks. With acts such as the EU AI Act the application of XAI in the biomedical domain becomes more and more essential. Our evaluation study represents a first step towards verifying the suitability of different XAI methods for various medical domains.

Hydrogel spacers (HS) are designed to minimise the radiation doses to the rectum in prostate cancer radiation therapy (RT) by creating a physical gap between the rectum and the target treatment volume inclusive of the prostate and seminal vesicles (SV). This study aims to determine the feasibility of incorporating diagnostic MRI (dMRI) information in statistical machine learning (SML) models developed with planning CT (pCT) anatomy for dose and rectal toxicity prediction. The SML models aim to support HS insertion decision-making prior to RT planning procedures. Regions of interest (ROIs) were retrospectively contoured on the pCT and registered dMRI scans for 20 patients. ROI Dice and Hausdorff distance (HD) comparison metrics were calculated. The ROI and patient clinical risk factors (CRFs) variables were inputted into three SML models and then pCT and dMRI-based dose and toxicity model performance compared through confusion matrices, AUC curves, accuracy performance metric results and observed patient outcomes. Average Dice values comparing dMRI and pCT ROIs were 0.81, 0.47 and 0.71 for the prostate, SV, and rectum respectively. Average Hausdorff distances were 2.15, 2.75 and 2.75 mm for the prostate, SV, and rectum respectively. The average accuracy metric across all models was 0.83 when using dMRI ROIs and 0.85 when using pCT ROIs. Differences between pCT and dMRI anatomical ROI variables did not impact SML model performance in this study, demonstrating the feasibility of using dMRI images. Due to the limited sample size further training of the predictive models including dMRI anatomy is recommended.

Polyps, like a silent time bomb in the gut, are always lurking and can explode into deadly colorectal cancer at any time. Many methods are attempted to maximize the early detection of colon polyps by screening, however, there are still face some challenges: (i) the scarcity of per-pixel annotation data and clinical features such as the blurred boundary and low contrast of polyps result in poor performance. (ii) existing weakly semi-supervised methods directly using pseudo-labels to supervise student tend to ignore the value brought by intermediate features in the teacher. To adapt the point-prompt teacher model to the challenging scenarios of complex medical images and limited annotation data, we creatively leverage the diverse inductive biases of CNN and Transformer to extract robust and complementary representation of polyp features (boundary and context). At the same time, a novel designed teacher-student intermediate feature distillation method is introduced rather than just using pseudo-labels to guide student learning. Comprehensive experiments demonstrate that our proposed method effectively handles scenarios with limited annotations and exhibits good segmentation performance. All code is available at https://github.com/dxqllp/WSS-Polyp.

Recent advancements in generative adversarial networks (GANs) have demonstrated substantial potential in medical image processing. Despite this progress, reconstructing images from incomplete data remains a challenge, impacting image quality. This systematic literature review explores the use of GANs in enhancing and reconstructing medical imaging data. A document survey of computing literature was conducted using the ACM Digital Library to identify relevant articles from journals and conference proceedings using keyword combinations, such as "generative adversarial networks or generative adversarial network," "medical image or medical imaging," and "image reconstruction." Across the reviewed articles, there were 122 datasets used in 175 instances, 89 top metrics employed 335 times, 10 different tasks with a total count of 173, 31 distinct organs featured in 119 instances, and 18 modalities utilized in 121 instances, collectively depicting significant utilization of GANs in medical imaging. The adaptability and efficacy of GANs were showcased across diverse medical tasks, organs, and modalities, utilizing top public as well as private/synthetic datasets for disease diagnosis, including the identification of conditions like cancer in different anatomical regions. The study emphasized GAN's increasing integration and adaptability in diverse radiology modalities, showcasing their transformative impact on diagnostic techniques, including cross-modality tasks. The intricate interplay between network size, batch size, and loss function refinement significantly impacts GAN's performance, although challenges in training persist. The study underscores GANs as dynamic tools shaping medical imaging, contributing significantly to image quality, training methodologies, and overall medical advancements, positioning them as substantial components driving medical advancements.

Research in the cross-modal medical image translation domain has been very productive over the past few years in tackling the scarce availability of large curated multi-modality datasets with the promising performance of GAN-based architectures. However, only a few of these studies assessed task-based related performance of these synthetic data, especially for the training of deep models. We design and compare different GAN-based frameworks for generating synthetic brain[18F]fluorodeoxyglucose (FDG) PET images from T1 weighted MRI data. We first perform standard qualitative and quantitative visual quality evaluation. Then, we explore further impact of using these fake PET data in the training of a deep unsupervised anomaly detection (UAD) model designed to detect subtle epilepsy lesions in T1 MRI and FDG PET images. We introduce novel diagnostic task-oriented quality metrics of the synthetic FDG PET data tailored to our unsupervised detection task, then use these fake data to train a use case UAD model combining a deep representation learning based on siamese autoencoders with a OC-SVM density support estimation model. This model is trained on normal subjects only and allows the detection of any variation from the pattern of the normal population. We compare the detection performance of models trained on 35 paired real MR T1 of normal subjects paired either on 35 true PET images or on 35 synthetic PET images generated from the best performing generative models. Performance analysis is conducted on 17 exams of epilepsy patients undergoing surgery. The best performing GAN-based models allow generating realistic fake PET images of control subject with SSIM and PSNR values around 0.9 and 23.8, respectively and in distribution (ID) with regard to the true control dataset. The best UAD model trained on these synthetic normative PET data allows reaching 74% sensitivity. Our results confirm that GAN-based models are the best suited for MR T1 to FDG PET translation, outperforming transformer or diffusion models. We also demonstrate the diagnostic value of these synthetic data for the training of UAD models and evaluation on clinical exams of epilepsy patients. Our code and the normative image dataset are available.

Computer-aided diagnosis systems based on deep neural networks heavily rely on datasets with high-quality labels. However, manual annotation for lesion diagnosis relies on image features, often requiring professional experience and complex image analysis process. This inevitably introduces noisy labels, which can misguide the training of classification models. Our goal is to design an effective method to address the challenges posed by label noise in medical images. we propose a novel noise-tolerant medical image classification framework consisting of two phases: fore-training correction and progressive hard-sample enhanced learning. In the first phase, we design a dual-branch sample partition detection scheme that effectively classifies each instance into one of three subsets: clean, hard, or noisy. Simultaneously, we propose a hard-sample label refinement strategy based on class prototypes with confidence-perception weighting and an effective joint correction method for noisy samples, enabling the acquisition of higher-quality training data. In the second phase, we design a progressive hard-sample reinforcement learning method to enhance the model's ability to learn discriminative feature representations. This approach accounts for sample difficulty and mitigates the effects of label noise in medical datasets. Our framework achieves an accuracy of 82.39% on the pneumoconiosis dataset collected by our laboratory. On a five-class skin disease dataset with six different levels of label noise (0, 0.05, 0.1, 0.2, 0.3, and 0.4), the average accuracy over the last ten epochs reaches 88.51%, 86.64%, 85.02%, 83.01%, 81.95%, 77.89%, respectively; For binary polyp classification under noise rates of 0.2, 0.3, and 0.4, the average accuracy over the last ten epochs is 97.90%, 93.77%, 89.33%, respectively. The effectiveness of our proposed framework is demonstrated through its performance on three challenging datasets with both real and synthetic noise. Experimental results further demonstrate the robustness of our method across varying noise rates.

The segmentation of breast cancer diagnosis and medical imaging contains issues such as noise, variation in contrast, and low resolutions which make it challenging to distinguish malignant sites. In this paper, we propose a new solution that integrates with AIPT (Advanced Image Preprocessing Techniques) and CASDN (Context-Aware Spatial Decomposition Network) to overcome these problems. The preprocessing pipeline apply bunch of methods including Adaptive Thresholding, Hierarchical Contrast Normalization, Contextual Feature Augmentation, Multi-Scale Region Enhancement, and Dynamic Histogram Equalization for image quality. These methods smooth edges, equalize the contrasting picture and inlay contextual details in a way which effectively eliminate the noise and make the images clearer and with fewer distortions. Experimental outcomes demonstrate its effectiveness by delivering a Dice Coefficient of 0.89, IoU of 0.85, and a Hausdorff Distance of 5.2 demonstrating its enhanced capability in segmenting significant tumor margins over other techniques. Furthermore, the use of the improved preprocessing pipeline benefits classification models with improved Convolutional Neural Networks having a classification accuracy of 85.3 % coupled with AUC-ROC of 0.90 which shows a significant enhancement from conventional techniques.•Enhanced segmentation accuracy with advanced preprocessing and CASDN, achieving superior performance metrics.•Robust multi-modality compatibility, ensuring effectiveness across mammograms, ultrasounds, and MRI scans.

Comprehending the biomechanical characteristics of the human lumbar spine is crucial for managing and preventing spinal disorders. Precise material properties derived from patient-specific CT scans are essential for simulations to accurately mimic real-life scenarios, which is invaluable in creating effective surgical plans. The integration of Finite Element Analysis (FEA) with Physics-Informed Neural Networks (PINNs) offers significant clinical benefits by automating lumbar spine segmentation and meshing. We developed a FEA model of the lumbar spine incorporating detailed anatomical and material properties derived from high-quality CT and MRI scans. The model includes vertebrae and intervertebral discs, segmented and meshed using advanced imaging and computational techniques. PINNs were implemented to integrate physical laws directly into the neural network training process, ensuring that the predictions of material properties adhered to the governing equations of mechanics. The model achieved an accuracy of 94.30% in predicting material properties such as Young's modulus (14.88 GPa for cortical bone and 1.23 MPa for intervertebral discs), Poisson's ratio (0.25 and 0.47, respectively), bulk modulus (9.87 GPa and 6.56 MPa, respectively), and shear modulus (5.96 GPa and 0.42 MPa, respectively). We developed a lumbar spine FEA model using anatomical and material properties from CT and MRI scans. Vertebrae and discs were segmented and meshed with advanced imaging techniques, while PINNs ensured material predictions followed mechanical laws. The integration of FEA and PINNs allows for accurate, automated prediction of material properties and mechanical behaviors of the lumbar spine, significantly reducing manual input and enhancing reliability. This approach ensures dependable biomechanical simulations and supports the development of personalized treatment plans and surgical strategies, ultimately improving clinical outcomes for spinal disorders. This method improves surgical planning and outcomes, contributing to better patient care and recovery in spinal disorders.

AI has emerged as a transformative force in clinical medicine, changing the diagnosis, treatment, and management of patients. Tools have been derived for working with ML, DL, and NLP algorithms to analyze large complex medical datasets with unprecedented accuracy and speed, thereby improving diagnostic precision, treatment personalization, and patient care outcomes. For example, CNNs have dramatically improved the accuracy of medical imaging diagnoses, and NLP algorithms have greatly helped extract insights from unstructured data, including EHRs. However, there are still numerous challenges that face AI integration into clinical workflows, including data privacy, algorithmic bias, ethical dilemmas, and problems with the interpretability of "black-box" AI models. These barriers have thus far prevented the widespread application of AI in health care, and its possible trends, obstacles, and future implications are necessary to be systematically explored. The purpose of this paper is, therefore, to assess the current trends in AI applications in clinical medicine, identify those obstacles that are hindering adoption, and identify possible future directions. This research hopes to synthesize evidence from other peer-reviewed articles to provide a more comprehensive understanding of the role that AI plays to advance clinical practices, improve patient outcomes, or enhance decision-making. A systematic review was done according to the PRISMA guidelines to explore the integration of Artificial Intelligence in clinical medicine, including trends, challenges, and future directions. PubMed, Cochrane Library, Web of Science, and Scopus databases were searched for peer-reviewed articles from 2014 to 2024 with keywords such as "Artificial Intelligence in Medicine," "AI in Clinical Practice," "Machine Learning in Healthcare," and "Ethical Implications of AI in Medicine." Studies focusing on AI application in diagnostics, treatment planning, and patient care reporting measurable clinical outcomes were included. Non-clinical AI applications and articles published before 2014 were excluded. Selected studies were screened for relevance, and then their quality was critically appraised to synthesize data reliably and rigorously. This systematic review includes the findings of 8 studies that pointed out the transformational role of AI in clinical medicine. AI tools, such as CNNs, had diagnostic accuracy more than the traditional methods, particularly in radiology and pathology. Predictive models efficiently supported risk stratification, early disease detection, and personalized medicine. Despite these improvements, significant hurdles, including data privacy, algorithmic bias, and resistance from clinicians regarding the "black-box" nature of AI, had yet to be surmounted. XAI has emerged as an attractive solution that offers the promise to enhance interpretability and trust. As a whole, AI appeared promising in enhancing diagnostics, treatment personalization, and clinical workflows by dealing with systemic inefficiencies. The transformation potential of AI in clinical medicine can transform diagnostics, treatment strategies, and efficiency. Overcoming obstacles such as concerns about data privacy, the danger of algorithmic bias, and difficulties with interpretability may pave the way for broader use and facilitate improvement in patient outcomes while transforming clinical workflows to bring sustainability into healthcare delivery.

In the HECKTOR 2022 challenge set [1], several state-of-the-art (SOTA, achieving best performance) deep learning models were introduced for predicting recurrence-free period (RFP) in head and neck cancer patients using PET and CT images. This study investigates whether a conventional DenseNet architecture, with optimized numbers of layers and image-fusion strategies, could achieve comparable performance as SOTA models. The HECKTOR 2022 dataset comprises 489 oropharyngeal cancer (OPC) patients from seven distinct centers. It was randomly divided into a training set (n = 369) and an independent test set (n = 120). Furthermore, an additional dataset of 400 OPC patients, who underwent chemo(radiotherapy) at our center, was employed for external testing. Each patients' data included pre-treatment CT- and PET-scans, manually generated GTV (Gross tumour volume) contours for primary tumors and lymph nodes, and RFP information. The present study compared the performance of DenseNet against three SOTA models developed on the HECKTOR 2022 dataset. When inputting CT, PET and GTV using the early fusion (considering them as different channels of input) approach, DenseNet81 (with 81 layers) obtained an internal test C-index of 0.69, a performance metric comparable with SOTA models. Notably, the removal of GTV from the input data yielded the same internal test C-index of 0.69 while improving the external test C-index from 0.59 to 0.63. Furthermore, compared to PET-only models, when utilizing the late fusion (concatenation of extracted features) with CT and PET, DenseNet81 demonstrated superior C-index values of 0.68 and 0.66 in both internal and external test sets, while using early fusion was better in only the internal test set. The basic DenseNet architecture with 81 layers demonstrated a predictive performance on par with SOTA models featuring more intricate architectures in the internal test set, and better performance in the external test. The late fusion of CT and PET imaging data yielded superior performance in the external test.