-
NPJ Precision Oncology Jun 2024Oral epithelial dysplasia (OED) is a premalignant histopathological diagnosis given to lesions of the oral cavity. Its grading suffers from significant...
Oral epithelial dysplasia (OED) is a premalignant histopathological diagnosis given to lesions of the oral cavity. Its grading suffers from significant inter-/intra-observer variability, and does not reliably predict malignancy progression, potentially leading to suboptimal treatment decisions. To address this, we developed an artificial intelligence (AI) algorithm, that assigns an Oral Malignant Transformation (OMT) risk score based on the Haematoxylin and Eosin (H&E) stained whole slide images (WSIs). Our AI pipeline leverages an in-house segmentation model to detect and segment both nuclei and epithelium. Subsequently, a shallow neural network utilises interpretable morphological and spatial features, emulating histological markers, to predict progression. We conducted internal cross-validation on our development cohort (Sheffield; n = 193 cases) and independent validation on two external cohorts (Birmingham and Belfast; n = 89 cases). On external validation, the proposed OMTscore achieved an AUROC = 0.75 (Recall = 0.92) in predicting OED progression, outperforming other grading systems (Binary: AUROC = 0.72, Recall = 0.85). Survival analyses showed the prognostic value of our OMTscore (C-index = 0.60, p = 0.02), compared to WHO (C-index = 0.64, p = 0.003) and binary grades (C-index = 0.65, p < 0.001). Nuclear analyses elucidated the presence of peri-epithelial and intra-epithelial lymphocytes in highly predictive patches of transforming cases (p < 0.001). This is the first study to propose a completely automated, explainable, and externally validated algorithm for predicting OED transformation. Our algorithm shows comparable-to-human-level performance, offering a promising solution to the challenges of grading OED in routine clinical practice.
PubMed: 38942998
DOI: 10.1038/s41698-024-00624-8 -
Nature Medicine Jun 2024As artificial intelligence (AI) rapidly approaches human-level performance in medical imaging, it is crucial that it does not exacerbate or propagate healthcare...
As artificial intelligence (AI) rapidly approaches human-level performance in medical imaging, it is crucial that it does not exacerbate or propagate healthcare disparities. Previous research established AI's capacity to infer demographic data from chest X-rays, leading to a key concern: do models using demographic shortcuts have unfair predictions across subpopulations? In this study, we conducted a thorough investigation into the extent to which medical AI uses demographic encodings, focusing on potential fairness discrepancies within both in-distribution training sets and external test sets. Our analysis covers three key medical imaging disciplines-radiology, dermatology and ophthalmology-and incorporates data from six global chest X-ray datasets. We confirm that medical imaging AI leverages demographic shortcuts in disease classification. Although correcting shortcuts algorithmically effectively addresses fairness gaps to create 'locally optimal' models within the original data distribution, this optimality is not true in new test settings. Surprisingly, we found that models with less encoding of demographic attributes are often most 'globally optimal', exhibiting better fairness during model evaluation in new test environments. Our work establishes best practices for medical imaging models that maintain their performance and fairness in deployments beyond their initial training contexts, underscoring critical considerations for AI clinical deployments across populations and sites.
PubMed: 38942996
DOI: 10.1038/s41591-024-03113-4 -
Scientific Reports Jun 2024In this study, in order to characterize the buried object via deep-learning-based surrogate modeling approach, 3-D full-wave electromagnetic simulations of a GPR model...
In this study, in order to characterize the buried object via deep-learning-based surrogate modeling approach, 3-D full-wave electromagnetic simulations of a GPR model have been used. The task is to independently predict characteristic parameters of a buried object of diverse radii allocated at different positions (depth and lateral position) in various dispersive subsurface media. This study has analyzed variable data structures (raw B-scans, extracted features, consecutive A-scans) with respect to computational cost and accuracy of surrogates. The usage of raw B-scan data and the applications for processing steps on B-scan profiles in the context of object characterization incur high computational cost so it can be a challenging issue. The proposed surrogate model referred to as the deep regression network (DRN) is utilized for time frequency spectrogram (TFS) of consecutive A-scans. DRN is developed with the main aim being computationally efficient (about 13 times acceleration) compared to conventional network models using B-scan images (2D data). DRN with TFS is favorably benchmarked to the state-of-the-art regression techniques. The experimental results obtained for the proposed model and second-best model, CNN-1D show mean absolute and relative error rates of 3.6 mm, 11.8 mm and 4.7%, 11.6% respectively. For the sake of supplementary verification under realistic scenarios, it is also applied for scenarios involving noisy data. Furthermore, the proposed surrogate modeling approach is validated using measurement data, which is indicative of suitability of the approach to handle physical measurements as data sources.
PubMed: 38942986
DOI: 10.1038/s41598-024-65996-0 -
Nature Aging Jun 2024Investigating the genetic underpinnings of human aging is essential for unraveling the etiology of and developing actionable therapies for chronic diseases. Here, we...
Investigating the genetic underpinnings of human aging is essential for unraveling the etiology of and developing actionable therapies for chronic diseases. Here, we characterize the genetic architecture of the biological age gap (BAG; the difference between machine learning-predicted age and chronological age) across nine human organ systems in 377,028 participants of European ancestry from the UK Biobank. The BAGs were computed using cross-validated support vector machines, incorporating imaging, physical traits and physiological measures. We identify 393 genomic loci-BAG pairs (P < 5 × 10) linked to the brain, eye, cardiovascular, hepatic, immune, metabolic, musculoskeletal, pulmonary and renal systems. Genetic variants associated with the nine BAGs are predominantly specific to the respective organ system (organ specificity) while exerting pleiotropic links with other organ systems (interorgan cross-talk). We find that genetic correlation between the nine BAGs mirrors their phenotypic correlation. Further, a multiorgan causal network established from two-sample Mendelian randomization and latent causal variance models revealed potential causality between chronic diseases (for example, Alzheimer's disease and diabetes), modifiable lifestyle factors (for example, sleep duration and body weight) and multiple BAGs. Our results illustrate the potential for improving human organ health via a multiorgan network, including lifestyle interventions and drug repurposing strategies.
PubMed: 38942983
DOI: 10.1038/s43587-024-00662-8 -
Scientific Reports Jun 2024In tuberculosis (TB), chest radiography (CXR) patterns are highly variable, mimicking pneumonia and many other diseases. This study aims to evaluate the efficacy of...
In tuberculosis (TB), chest radiography (CXR) patterns are highly variable, mimicking pneumonia and many other diseases. This study aims to evaluate the efficacy of Google teachable machine, a deep neural network-based image classification tool, to develop algorithm for predicting TB probability of CXRs. The training dataset included 348 TB CXRs and 3806 normal CXRs for training TB detection. We also collected 1150 abnormal CXRs and 627 normal CXRs for training abnormality detection. For external validation, we collected 250 CXRs from our hospital. We also compared the accuracy of the algorithm to five pulmonologists and radiological reports. In external validation, the AI algorithm showed areas under the curve (AUC) of 0.951 and 0.975 in validation dataset 1 and 2. The accuracy of the pulmonologists on validation dataset 2 showed AUC range of 0.936-0.995. When abnormal CXRs other than TB were added, AUC decreased in both human readers (0.843-0.888) and AI algorithm (0.828). When combine human readers with AI algorithm, the AUC further increased to 0.862-0.885. The TB CXR AI algorithm developed by using Google teachable machine in this study is effective, with the accuracy close to experienced clinical physicians, and may be helpful for detecting tuberculosis by CXR.
Topics: Humans; Deep Learning; Tuberculosis, Pulmonary; Radiography, Thoracic; Algorithms; Female; Male; Middle Aged; Adult; Area Under Curve
PubMed: 38942819
DOI: 10.1038/s41598-024-65703-z -
Scientific Reports Jun 2024Prostate cancer is one of the most common and fatal diseases among men, and its early diagnosis can have a significant impact on the treatment process and prevent...
Prostate cancer is one of the most common and fatal diseases among men, and its early diagnosis can have a significant impact on the treatment process and prevent mortality. Since it does not have apparent clinical symptoms in the early stages, it is difficult to diagnose. In addition, the disagreement of experts in the analysis of magnetic resonance images is also a significant challenge. In recent years, various research has shown that deep learning, especially convolutional neural networks, has appeared successfully in machine vision (especially in medical image analysis). In this research, a deep learning approach was used on multi-parameter magnetic resonance images, and the synergistic effect of clinical and pathological data on the accuracy of the model was investigated. The data were collected from Trita Hospital in Tehran, which included 343 patients (data augmentation and learning transfer methods were used during the process). In the designed model, four different types of images are analyzed with four separate ResNet50 deep convolutional networks, and their extracted features are transferred to a fully connected neural network and combined with clinical and pathological features. In the model without clinical and pathological data, the maximum accuracy reached 88%, but by adding these data, the accuracy increased to 96%, which shows the significant impact of clinical and pathological data on the accuracy of diagnosis.
Topics: Humans; Deep Learning; Prostatic Neoplasms; Male; Middle Aged; Aged; Neural Networks, Computer; Magnetic Resonance Imaging; Multiparametric Magnetic Resonance Imaging; Image Processing, Computer-Assisted; Image Interpretation, Computer-Assisted; Iran
PubMed: 38942817
DOI: 10.1038/s41598-024-65354-0 -
Scientific Reports Jun 2024This study aimed to develop and validate a machine learning (ML) model tailored to the Korean population with type 2 diabetes mellitus (T2DM) to provide a superior...
This study aimed to develop and validate a machine learning (ML) model tailored to the Korean population with type 2 diabetes mellitus (T2DM) to provide a superior method for predicting the development of cardiovascular disease (CVD), a major chronic complication in these patients. We used data from two cohorts, namely the discovery (one hospital; n = 12,809) and validation (two hospitals; n = 2019) cohorts, recruited between 2008 and 2022. The outcome of interest was the presence or absence of CVD at 3 years. We selected various ML-based models with hyperparameter tuning in the discovery cohort and performed area under the receiver operating characteristic curve (AUROC) analysis in the validation cohort. CVD was observed in 1238 (10.2%) patients in the discovery cohort. The random forest (RF) model exhibited the best overall performance among the models, with an AUROC of 0.830 (95% confidence interval [CI] 0.818-0.842) in the discovery dataset and 0.722 (95% CI 0.660-0.783) in the validation dataset. Creatinine and glycated hemoglobin levels were the most influential factors in the RF model. This study introduces a pioneering ML-based model for predicting CVD in Korean patients with T2DM, outperforming existing prediction tools and providing a groundbreaking approach for early personalized preventive medicine.
Topics: Humans; Cardiovascular Diseases; Machine Learning; Diabetes Mellitus, Type 2; Female; Male; Middle Aged; Republic of Korea; Aged; Cohort Studies; ROC Curve; Risk Factors
PubMed: 38942775
DOI: 10.1038/s41598-024-63798-y -
Scientific Reports Jun 2024A novel interval valued p,q Rung orthopair fuzzy (IVPQ-ROF) multiple attribute group decision making (MAGDM) method for sustainable supplier selection (SSS) is proposed...
A novel interval valued p,q Rung orthopair fuzzy (IVPQ-ROF) multiple attribute group decision making (MAGDM) method for sustainable supplier selection (SSS) is proposed in this paper. This study mainly contains two research points: (1) tackling the interrelation between attributes; and (2) describing the psychological state and risk attitude of decision makers (DMs). For the first research point, we introduce the Archimedean operation rules for interval valued p,q Rung orthopair fuzzy sets (IVPQ-ROFSs), then the generalized interval valued p, q Rung orthopair fuzzy Maclaurin symmetric mean (GIVPQ-ROFMSM) operator and the generalized interval valued p, q Rung orthopair fuzzy weighted Maclaurin symmetric mean (GIVPQ-ROFWMSM) operator are defined to reflect the correlation between attributes. For the second research point, we introduce the positive ideal degree (PID) and negative ideal degree (NID) based on projection of IVPQ-ROFSs, and modified regret theory. Both of them consider the best alternative and worst alternative, so as to reflect the psychological state and risk attitude of DMs. Finally, a SSS problem is presented to manifest the effectiveness of the designed method. We also provide sensitivity analysis and comparative analysis to further demonstrate the rationality and validity of the proposed method.
PubMed: 38942771
DOI: 10.1038/s41598-024-64765-3 -
Journal of Chemical Theory and... Jun 2024The optimal interaction of drugs with plasma membranes and membranes of subcellular organelles is a prerequisite for desirable pharmacology. Importantly, for drugs...
Application of Generative Artificial Intelligence in Predicting Membrane Partitioning of Drugs: Combining Denoising Diffusion Probabilistic Models and MD Simulations Reduces the Computational Cost to One-Third.
The optimal interaction of drugs with plasma membranes and membranes of subcellular organelles is a prerequisite for desirable pharmacology. Importantly, for drugs targeting the transmembrane lipid-facing sites of integral membrane proteins, the relative affinity of a drug to the bilayer lipids compared to the surrounding aqueous phase affects the partitioning, access, and binding of the drug to the target site. Molecular dynamics (MD) simulations, including enhanced sampling techniques such as steered MD, umbrella sampling (US), and metadynamics, offer valuable insights into the interactions of drugs with the membrane lipids and water in atomistic detail. However, these methods are computationally prohibitive for the high-throughput screening of drug candidates. This study shows that applying denoising diffusion probabilistic models (DDPMs), a generative AI method, to US simulation data reduces the computational cost significantly. Specifically, the models used only partial (one-third) data from the US simulations and reproduced the complete potential of mean force (PMF) profiles for three FDA-approved drugs (β2-adrenergic agonists) and ∼20 biologically relevant chemicals with known experimentally characterized bilayer locations. Intriguingly, the model can predict the solvation-free energies for partitioning and crossing the bilayer, preferred bilayer locations (low-energy well), and orientations of the ligands with high accuracy. The results indicate that DDPMs can be used to characterize the complete membrane partitioning profile of drug molecules using fewer umbrella sampling simulations at select positions along the bilayer normal (-axis), irrespective of their amphiphilic-lipophilic-cephalophilic characteristics.
PubMed: 38942732
DOI: 10.1021/acs.jctc.4c00315 -
Clinical Radiology May 2024In the rapidly evolving field of artificial intelligence (AI) for radiology, with a plethora of vendor options and use-cases and evidence claims to sift through, the... (Review)
Review
In the rapidly evolving field of artificial intelligence (AI) for radiology, with a plethora of vendor options and use-cases and evidence claims to sift through, the pressing question is how to effectively implement the right tool for enhanced patient care? This article presents a structured approach to AI deployment, drawing from a comprehensive case study in South West London. We underscore the necessity of forming a dedicated AI team with a clear vision and assertive leadership to navigate such complexities. Central to our discussion is the significance of crafting an AI implementation plan, with an overarching aim to augment patient care, promote operational efficiency, and lay down standardized protocols for seamless AI adoption. By presenting a blueprint for AI implementation within the National Health Service (NHS), we intend to demystify the process for radiology departments across the UK, enabling them to make informed decisions and empowering their staff to embrace and leverage AI responsibly ensuring that patient welfare remains at the heart of innovation. Thus, having a framework to follow when implementing an AI solution that addresses a vision for scalable adoption, core team members with diversity of skillset, staff engagement and education, plan for vendor selection, and change management is crucial for success.
PubMed: 38942706
DOI: 10.1016/j.crad.2024.05.018