Exploring the early stages of Alzheimer's disease (AD) is crucial for timely intervention to help manage symptoms and set expectations for affected individuals and their families. However, the study of the early stages of AD involves analysing heterogeneous disease cohorts which may present challenges for some modelling techniques. This heterogeneity stems from the diverse nature of AD itself, as well as the inclusion of undiagnosed or 'at-risk' AD individuals or the presence of comorbidities which differentially affect AD biomarkers within the cohort. Normative modelling is an emerging technique for studying heterogeneous disorders that can quantify how brain imaging-based measures of individuals deviate from a healthy population. The normative model provides a statistical description of the 'normal' range that can be used at subject level to detect deviations, which may relate to pathological effects. In this work, we applied a deep learning-based normative model, pre-trained on MRI scans in the UK Biobank, to investigate ageing and identify abnormal age-related decline. We calculated deviations, relative to the healthy population, in multi-modal MRI data of non-demented individuals in the external EPAD (ep-ad.org) cohort and explored these deviations with the aim of determining whether normative modelling could detect AD-relevant subtle differences between individuals. We found that aggregate measures of deviation based on the entire brain correlated with measures of cognitive ability and biological phenotypes, indicating the effectiveness of a general deviation metric in identifying AD-related differences among individuals. We then explored deviations in individual imaging features, stratified by cognitive performance and genetic risk, across different brain regions and found that the brain regions showing deviations corresponded to those affected by AD such as the hippocampus. Finally, we found that 'at-risk' individuals in the EPAD cohort exhibited increasing deviation over time, with an approximately 6.4 times greater t-statistic in a pairwise t-test compared to a 'super-healthy' cohort. This study highlights the capability of deep normative modelling approaches to detect subtle differences in brain morphology among individuals at risk of developing AD in a non-demented population. Our findings allude to the potential utility of normative deviation metrics in monitoring disease progression.

Only in recent years it has been demonstrated that the thoracolumbar fascia is involved in low back pain (LBP), thus highlighting its implications for treatments. Furthermore, an easily accessible and non-invasive way to investigate the fascia in real time is the ultrasound examination, which to be reliable as is, it must overcome the challenges related to the configuration of the machine and the experience of the operator. Therefore, the lack of a clear understanding of the fascial system combined with the penalty related to the setting of the ultrasound acquisition has generated a gap that makes its effective evaluation difficult during clinical routine. The aim of the present work is to fill this gap by investigating the effectiveness of using a deep learning approach to segment the thoracolumbar fascia from ultrasound imaging. A total of 538 ultrasound images of the thoracolumbar fascia of LBP subjects were finally used to train and test a deep learning network. An additional test set (so-called Test set 2) was collected from another center, operator, machine manufacturer, patient cohort, and protocol to improve the generalizability of the study. A U-Net-based architecture was demonstrated to be able to segment these structures with a final training accuracy of 0.99 and a validation accuracy of 0.91. The accuracy of the prediction computed on a test set (87 images not included in the training set) reached the 0.94, with a mean intersection over union index of 0.82 and a Dice-score of 0.76. These latter metrics were outperformed by those in Test set 2. The validity of the predictions was also verified and confirmed by two expert clinicians. Automatic identification of the thoracolumbar fascia has shown promising results to thoroughly investigate its alteration and target a personalized rehabilitation intervention based on each patient-specific scenario.

Due to differences in subjective experience and professional level among doctors, as well as inconsistent diagnostic criteria, there are issues with the accuracy and reliability of single imaging diagnosis results for knee joint injuries. To address these issues, magnetic resonance imaging (MRI), computed tomography (CT) and ultrasound (US) are adopted in this article for ensemble learning, and deep learning (DL) is combined for automatic analysis. By steps such as image enhancement, noise elimination, and tissue segmentation, the quality of image data is improved, and then convolutional neural networks (CNN) are used to automatically identify and classify injury types. The experimental results show that the DL model exhibits high sensitivity and specificity in the diagnosis of different types of injuries, such as anterior cruciate ligament tear, meniscus injury, cartilage injury, and fracture. The diagnostic accuracy of anterior cruciate ligament tear exceeds 90%, and the highest diagnostic accuracy of cartilage injury reaches 95.80%. In addition, compared with traditional manual image interpretation, the DL model has significant advantages in time efficiency, with a significant reduction in average interpretation time per case. The diagnostic consistency experiment shows that the DL model has high consistency with doctors' diagnosis results, with an overall error rate of less than 2%. The model has high accuracy and strong generalization ability when dealing with different types of joint injuries. These data indicate that combining multiple imaging technologies and the DL algorithm can effectively improve the accuracy and efficiency of diagnosing sports injuries of knee joints.

Chemical exchange saturation transfer (CEST) magnetic resonance imaging (MRI) has emerged as a powerful tool to image endogenous or exogenous macromolecules. CEST contrast highly depends on radiofrequency irradiation B₁ level. Spatial inhomogeneity of B₁ field would bias CEST measurement. Conventional interpolation-based B₁ correction method required CEST dataset acquisition under multiple B₁ levels, substantially prolonging scan time. The recently proposed supervised deep learning approach reconstructed B₁ inhomogeneity corrected CEST effect at the identical B₁ as of the training data, hindering its generalization to other B₁ levels. In this study, we proposed a Conditional Variational Autoencoder (CVAE)-based generative model to generate B₁ inhomogeneity corrected Z spectra from single CEST acquisition. The model was trained from pixel-wise source-target paired Z spectra under multiple B₁ with target B₁ as a conditional variable. Numerical simulation and healthy human brain imaging at 5 T were respectively performed to evaluate the performance of proposed model in B₁ inhomogeneity corrected CEST MRI. Results showed that the generated B₁-corrected Z spectra agreed well with the reference averaged from regions with subtle B₁ inhomogeneity. Moreover, the performance of the proposed model in correcting B₁ inhomogeneity in APT CEST effect, as measured by both MTR_asym and [Formula: see text] at 3.5 ppm, were superior over conventional Z/contrast-B₁-interpolation and other deep learning methods, especially when target B₁ were not included in sampling or training dataset. In summary, the proposed model allows generalized B₁ inhomogeneity correction, benefiting quantitative CEST MRI in clinical routines.

Pediatric bipolar disorder (PBD) is a highly debilitating condition, characterized by alternating episodes of mania and depression, with intervening periods of remission. Limited information is available about the functional and structural abnormalities in PBD, particularly when comparing type I with type II subtypes. Resting-state brain activity and structural grey matter, assessed through MRI, may provide insight into the neurobiological biomarkers of this disorder. In this study, Resting state Regional Homogeneity (ReHo) and grey matter concentration (GMC) data of 58 PBD patients, and 21 healthy controls matched for age, gender, education and IQ, were analyzed in a data fusion unsupervised machine learning approach known as transposed Independent Vector Analysis. Two networks significantly differed between BPD and HC. The first network included fronto- medial regions, such as the medial and superior frontal gyrus, the cingulate, and displayed higher ReHo and GMC values in PBD compared to HC. The second network included temporo-posterior regions, as well as the insula, the caudate and the precuneus and displayed lower ReHo and GMC values in PBD compared to HC. Additionally, two networks differ between type-I vs type-II in PBD: an occipito-cerebellar network with increased ReHo and GMC in type-I compared to type-II, and a fronto-parietal network with decreased ReHo and GMC in type-I compared to type-II. Of note, the first network positively correlated with depression scores. These findings shed new light on the functional and structural abnormalities displayed by pediatric bipolar patients.

Osteoarthritis (OA) is a major cause of disability worldwide, with symptoms like joint pain, limited functionality, and decreased quality of life, potentially leading to deformity and irreversible damage. Chemical changes in joint tissues precede imaging alterations, making early diagnosis challenging for conventional methods like X-rays. Although Raman imaging provides detailed chemical information, it is time-consuming. This paper aims to achieve rapid osteoarthritis diagnosis and grading using a self-developed Raman imaging system combined with deep learning denoising and acceleration algorithms. Our self-developed aberration-corrected line-scanning confocal Raman imaging device acquires a line of Raman spectra (hundreds of points) per scan using a galvanometer or displacement stage, achieving spatial and spectral resolutions of 2 μm and 0.2 nm, respectively. Deep learning algorithms enhance the imaging speed by over 4 times through effective spectrum denoising and signal-to-noise ratio (SNR) improvement. By leveraging the denoising capabilities of deep learning, we are able to acquire high-quality Raman spectral data with a reduced integration time, thereby accelerating the imaging process. Experiments on the tibial plateau of osteoarthritis patients compared three excitation wavelengths (532, 671, and 785 nm), with 671 nm chosen for optimal SNR and minimal fluorescence. Machine learning algorithms achieved a 98 % accuracy in distinguishing articular from calcified cartilage and a 97 % accuracy in differentiating osteoarthritis grades I to IV. Our fast Raman imaging system, combining an aberration-corrected line-scanning confocal Raman imager with deep learning denoising, offers improved imaging speed and enhanced spectral and spatial resolutions. It enables rapid, label-free detection of osteoarthritis severity and can identify early compositional changes before clinical imaging, allowing precise grading and tailored treatment, thus advancing orthopedic diagnostics and improving patient outcomes.

Deep learning has been widely used in research on sparse-view computed tomography (CT) image reconstruction. While sufficient training data can lead to high accuracy, collecting medical images is often challenging due to legal or ethical concerns, making it necessary to develop methods that perform well with limited data. To address this issue, we explored the use of nonmedical images for pre-training. Therefore, in this study, we investigated whether fractal images could improve the quality of sparse-view CT images, even with a reduced number of medical images. Fractal images generated by an iterated function system (IFS) were used for nonmedical images, and medical images were obtained from the CHAOS dataset. Sinograms were then generated using 36 projections in sparse-view and the images were reconstructed by filtered back-projection (FBP). FBPConvNet and WNet (first module: learning fractal images, second module: testing medical images, and third module: learning output) were used as networks. The effectiveness of pre-training was then investigated for each network. The quality of the reconstructed images was evaluated using two indices: structural similarity (SSIM) and peak signal-to-noise ratio (PSNR). The network parameters pre-trained with fractal images showed reduced artifacts compared to the network trained exclusively with medical images, resulting in improved SSIM. WNet outperformed FBPConvNet in terms of PSNR. Pre-training WNet with fractal images produced the best image quality, and the number of medical images required for main-training was reduced from 5000 to 1000 (80% reduction). Using fractal images for network training can reduce the number of medical images required for artifact reduction in sparse-view CT. Therefore, fractal images can improve accuracy even with a limited amount of training data in deep learning.

Artificial intelligence (AI) studies show promise in enhancing accuracy and efficiency in mammographic screening programs worldwide. However, its integration into clinical workflows faces several challenges, including unintended errors, the need for professional training, and ethical concerns. Notably, specific frameworks for AI imaging in breast cancer screening are still lacking. This study aims to identify the challenges associated with implementing AI in breast screening programs and to apply the Consolidated Framework for Implementation Research (CFIR) to discuss a practical governance framework for AI in this context. Three electronic databases (PubMed, Embase, and MEDLINE) were searched using combinations of the keywords "artificial intelligence," "regulation," "governance," "breast cancer," and "screening." Original studies evaluating AI in breast cancer detection or discussing challenges related to AI implementation in this setting were eligible for review. Findings were narratively synthesized and subsequently mapped directly onto the constructs within the CFIR. A total of 1240 results were retrieved, with 20 original studies ultimately included in this systematic review. The majority (n=19) focused on AI-enhanced mammography, while 1 addressed AI-enhanced ultrasound for women with dense breasts. Most studies originated from the United States (n=5) and the United Kingdom (n=4), with publication years ranging from 2019 to 2023. The quality of papers was rated as moderate to high. The key challenges identified were reproducibility, evidentiary standards, technological concerns, trust issues, as well as ethical, legal, societal concerns, and postadoption uncertainty. By aligning these findings with the CFIR constructs, action plans targeting the main challenges were incorporated into the framework, facilitating a structured approach to addressing these issues. This systematic review identifies key challenges in implementing AI in breast cancer screening, emphasizing the need for consistency, robust evidentiary standards, technological advancements, user trust, ethical frameworks, legal safeguards, and societal benefits. These findings can serve as a blueprint for policy makers, clinicians, and AI developers to collaboratively advance AI adoption in breast cancer screening. PROSPERO CRD42024553889; https://tinyurl.com/mu4nwcxt.

Chronic obstructive pulmonary disease (COPD) is a common and progressive respiratory condition characterized by persistent airflow limitation and symptoms such as dyspnea, cough, and sputum production. Acute exacerbations (AE) of COPD (AE-COPD) are key determinants of disease progression; yet, existing predictive models relying mainly on spirometric measurements, such as forced expiratory volume in 1 second, reflect only a fraction of the physiological information embedded in respiratory function tests. Recent advances in artificial intelligence (AI) have enabled more sophisticated analyses of full spirometric curves, including flow-volume loops and volume-time curves, facilitating the identification of complex patterns associated with increased exacerbation risk. This study aimed to determine whether a predictive model that integrates clinical data and spirometry images with the use of AI improves accuracy in predicting moderate-to-severe and severe AE-COPD events compared to a clinical-only model. A retrospective cohort study was conducted using COPD registry data from 2 teaching hospitals from January 2004 to December 2020. The study included a total of 10,492 COPD cases, divided into a development cohort (6870 cases) and an external validation cohort (3622 cases). The AI-enhanced model (AI-PFT-Clin) used a combination of clinical variables (eg, history of AE-COPD, dyspnea, and inhaled treatments) and spirometry image data (flow-volume loop and volume-time curves). In contrast, the Clin model used only clinical variables. The primary outcomes were moderate-to-severe and severe AE-COPD events within a year of spirometry. In the external validation cohort, the AI-PFT-Clin model outperformed the Clin model, showing an area under the receiver operating characteristic curve of 0.755 versus 0.730 (P<.05) for moderate-to-severe AE-COPD and 0.713 versus 0.675 (P<.05) for severe AE-COPD. The AI-PFT-Clin model demonstrated reliable predictive capability across subgroups, including younger patients and those without previous exacerbations. Higher AI-PFT-Clin scores correlated with elevated AE-COPD risk (adjusted hazard ratio for Q4 vs Q1: 4.21, P<.001), with sustained predictive stability over a 10-year follow-up period. The AI-PFT-Clin model, by integrating clinical data with spirometry images, offers enhanced predictive accuracy for AE-COPD events compared to a clinical-only approach. This AI-based framework facilitates the early identification of high-risk individuals through the detection of physiological abnormalities not captured by conventional metrics. The model's robust performance and long-term predictive stability suggest its potential utility in proactive COPD management and personalized intervention planning. These findings highlight the promise of incorporating advanced AI techniques into routine COPD management, particularly in populations traditionally seen as lower risk, supporting improved management of COPD through tailored patient care.

Medical image segmentation models are often trained on curated datasets, leading to performance degradation when deployed in real-world clinical settings due to mismatches between training and test distributions. While data augmentation techniques are widely used to address these challenges, traditional visually consistent augmentation strategies lack the robustness needed for diverse real-world scenarios. In this work, we systematically evaluate alternative augmentation strategies, focusing on MixUp and Auxiliary Fourier Augmentation. These methods mitigate the effects of multiple variations without explicitly targeting specific sources of distribution shifts. We demonstrate how these techniques significantly improve out-of-distribution generalization and robustness to imaging variations across a wide range of transformations in cardiac cine MRI and prostate MRI segmentation. We quantitatively find that these augmentation methods enhance learned feature representations by promoting separability and compactness. Additionally, we highlight how their integration into nnU-Net training pipelines provides an easy-to-implement, effective solution for enhancing the reliability of medical segmentation models in real-world applications.