Talanta. 2024 Dec 2;285:127268. doi: 10.1016/j.talanta.2024.127268. Online ahead of print.

ABSTRACT

Foodborne pathogenic infections pose a significant threat to human health. Accurate detection of foodborne diseases is essential in preventing disease transmission. This study proposed an AI model for precisely identifying foodborne pathogenic bacteria based on single-cell Raman spectrum. Self-transfer deep learning and ensemble prediction algorithms had been incorporated into the model framework to improve training efficiency and predictive performance, significantly improving prediction results. Our model can identify simultaneously gram-negative and positive, genus, species of foodborne pathogenic bacteria with an accuracy over 99.99 %, as well as recognized strain with over 99.49 %. At all four classification levels, unprecedented excellent predictive performance had been achieved. This advancement holds practical significance for medical detection and diagnosis of foodborne diseases by reducing false negatives.

PMID:39644671 | DOI:10.1016/j.talanta.2024.127268

Neural Netw. 2024 Dec 4;183:106983. doi: 10.1016/j.neunet.2024.106983. Online ahead of print.

ABSTRACT

Stability analysis is an essential aspect of studying the generalization ability of deep learning, as it involves deriving generalization bounds for stochastic gradient descent-based training algorithms. Adversarial training is the most widely used defense against adversarial attacks. However, previous generalization bounds for adversarial training have not included information regarding data distribution. In this paper, we fill this gap by providing generalization bounds for stochastic gradient descent-based adversarial training that incorporate data distribution information. We utilize the concepts of on-average stability and high-order approximate Lipschitz conditions to examine how changes in data distribution and adversarial budget can affect robust generalization gaps. Our derived generalization bounds for both convex and non-convex losses are at least as good as the uniform stability-based counterparts which do not include data distribution information. Furthermore, our findings demonstrate how distribution shifts from data poisoning attacks can impact robust generalization.

PMID:39644596 | DOI:10.1016/j.neunet.2024.106983

Comput Biol Med. 2024 Dec 6;185:109449. doi: 10.1016/j.compbiomed.2024.109449. Online ahead of print.

ABSTRACT

This survey paper provides a comprehensive analysis of various Deep Learning (DL) techniques and algorithms for detecting protein-protein interactions (PPIs). It examines the scalability, interpretability, accuracy, and efficiency of each technique, offering a detailed empirical and experimental evaluation. Empirically, the techniques are assessed based on four key criteria, while experimentally, they are ranked by specific algorithms and broader methodological categories. Deep Neural Networks (DNNs) demonstrated high accuracy but faced limitations such as overfitting and low interpretability. Convolutional Neural Networks (CNNs) were highly efficient at extracting hierarchical features from biological sequences, while Generative Stochastic Networks (GSNs) excelled in handling uncertainty. Long Short-Term Memory (LSTM) networks effectively captured temporal dependencies within PPI sequences, though they presented scalability challenges. This paper concludes with insights into potential improvements and future directions for advancing DL techniques in PPI identification, highlighting areas where further optimization can enhance performance and applicability.

PMID:39644584 | DOI:10.1016/j.compbiomed.2024.109449

Comput Biol Med. 2024 Dec 6;185:109500. doi: 10.1016/j.compbiomed.2024.109500. Online ahead of print.

ABSTRACT

Emphysema, marked by irreversible lung tissue destruction, poses challenges in progression prediction due to its heterogeneity. Early detection is particularly critical for patients with Alpha-1 Antitrypsin Deficiency (AATD), a genetic disorder reducing ATT protein levels. Heterozygous carriers (PiMS and PiMZ) have variable AAT levels thus complicating their prognosis. This study introduces a novel prognostic model, the Lobe-based Transformer encoder (LobTe), designed to predict the annual change in lung density (ΔALD [g/L-yr]) using CT scans. Utilizing a global self-attention mechanism, LobTe specifically analyzes lobar tissue destruction to forecast disease progression. In parallel, we developed and compared a second model utilizing an LSTM architecture that implements a local subject-specific attention mechanism. Our methodology was validated on a cohort of 2,019 participants from the COPDGene study. The LobTe model demonstrated a small root mean squared error (RMSE=1.73 g/L-yr) and a notable correlation coefficient (ρ=0.61), explaining over 35% of the variability in ΔALD (R2= 0.36). Notably, it achieved a higher correlation coefficient of 0.68 for PiMZ heterozygous carriers, indicating its effectiveness in detecting early emphysema progression among smokers with mild to moderate AAT deficiency. The presented models could serve as a tool for monitoring disease progression and informing treatment strategies in carriers and subjects with AATD. Our code is available at github.com/acil-bwh/LobTe.

PMID:39644582 | DOI:10.1016/j.compbiomed.2024.109500

J Comput Chem. 2025 Jan 5;46(1):e27518. doi: 10.1002/jcc.27518.

ABSTRACT

Predicting drug target binding affinity has huge relevance in Modern drug discovery and drug repositioning processes which assist doctors to come up with new drugs or even use the existing drugs for new target proteins. In silico models, using advanced deep learning techniques could further assist these prediction tasks by providing most prominent drug target pairs. Considering these factors, a deep learning based algorithmic framework is developed in this study to support drug target interaction prediction. The proposed SeqDPI model extract the relevant drug and protein features from the one dimensional Sequential representation of the dataset considered using optimized CNN networks that deploy convolutions on varying length of amino acid subsequence's to capture hidden pattern, the convolved drug- protein features obtained are then used as an input to L2 penalized feed forward neural network which matches the local residue patterns in protein classes with molecular fingerprints of drugs to predict the binding strength for all drug target pairs. The proposed model reduces the convolution strain typically encountered in existing in silico models that utilize complex 3D structures of drug protein datasets. The result shows that the SeqDPI model achieves a mean square error MSE of (0.167) across cross validation folds, outperforming baseline models such as KronRLS (0.406), Simboost (0.226), and DeepPS (0.214). Additionally, SeqDPI attains a high CI score of 0.9114 on the benchmark KIBA dataset, demonstrating its statistical significance and computational efficiency compared to existing methods. This gives the relevance and effectiveness of SeqDPI model in accurately predicting binding affinities while working with simpler one-dimensional data, making it a robust and computationally cost-effective solution for drug-target interaction prediction.

PMID:39644133 | DOI:10.1002/jcc.27518

Environ Monit Assess. 2024 Dec 7;197(1):37. doi: 10.1007/s10661-024-13468-3.

ABSTRACT

In the agricultural sector, sugarcane farming is one of the most organized forms of cultivation. India is the second-largest producer of sugarcane in the world. However, sugarcane crops are highly affected by diseases, which significantly affect crop production. Despite development in deep learning techniques, disease detection remains a challenging and time-consuming task. This paper presents a novel Hybrid Shifted-Vision Transformer approach for the automated classification of sugarcane leaf diseases. The model integrates the Vision Transformer architecture with Hybrid Shifted Windows to effectively capture both local and global features, which is crucial for accurately identifying disease patterns at different spatial scales. To improve feature representation and model performance, self-supervised learning is employed using data augmentation techniques like random rotation, flipping, and occlusion, combined with a jigsaw puzzle task that helps the model learn spatial relationships in images. The method addresses class imbalances in the dataset through stratified sampling, ensuring balanced training and testing sets. The approach is fine-tuned on sugarcane leaf disease datasets using categorical cross-entropy loss, minimizing dissimilarity between predicted probabilities and real labels. Experimental results demonstrate that the Hybrid Shifted-Vision Transformer outperforms traditional models, achieving higher accuracy in disease detection of 98.5%, making it crucial for reliable disease diagnosis and decision-making in agriculture. This architecture enables efficient, large-scale automated sugarcane disease monitoring.

PMID:39643787 | DOI:10.1007/s10661-024-13468-3

Arch Bronconeumol. 2024 Nov 22:S0300-2896(24)00439-3. doi: 10.1016/j.arbres.2024.11.001. Online ahead of print.

ABSTRACT

In this narrative review, we address the ongoing challenges of lung cancer (LC) screening using chest low-dose computerized tomography (LDCT) and explore the contributions of artificial intelligence (AI), in overcoming them. We focus on evaluating the initial (baseline) LDCT examination, which provides a wealth of information relevant to the screening participant's health. This includes the detection of large-size prevalent LC and small-size malignant nodules that are typically diagnosed as LCs upon growth in subsequent annual LDCT scans. Additionally, the baseline LDCT examination provides valuable information about smoking-related comorbidities, including cardiovascular disease, chronic obstructive pulmonary disease, and interstitial lung disease (ILD), by identifying relevant markers. Notably, these comorbidities, despite the slow progression of their markers, collectively exceed LC as ultimate causes of death at follow-up in LC screening participants. Computer-assisted diagnosis tools currently improve the reproducibility of radiologic readings and reduce the false negative rate of LDCT. Deep learning (DL) tools that analyze the radiomic features of lung nodules are being developed to distinguish between benign and malignant nodules. Furthermore, AI tools can predict the risk of LC in the years following a baseline LDCT. AI tools that analyze baseline LDCT examinations can also compute the risk of cardiovascular disease or death, paving the way for personalized screening interventions. Additionally, DL tools are available for assessing osteoporosis and ILD, which helps refine the individual's current and future health profile. The primary obstacles to AI integration into the LDCT screening pathway are the generalizability of performance and the explainability.

PMID:39643515 | DOI:10.1016/j.arbres.2024.11.001

J Magn Reson Imaging. 2024 Dec 6. doi: 10.1002/jmri.29670. Online ahead of print.

ABSTRACT

BACKGROUND: Previous studies explored MRI-based radiomic features for differentiating between human epidermal growth factor receptor 2 (HER2)-zero, HER2-low, and HER2-positive breast cancer, but deep learning's effectiveness is uncertain.

PURPOSE: This study aims to develop and validate a deep learning system using dynamic contrast-enhanced MRI (DCE-MRI) for automated tumor segmentation and classification of HER2-zero, HER2-low, and HER2-positive statuses.

STUDY TYPE: Retrospective.

POPULATION: One thousand two hundred ninety-four breast cancer patients from three centers who underwent DCE-MRI before surgery were included in the study (52 ± 11 years, 811/204/279 for training/internal testing/external testing).

FIELD STRENGTH/SEQUENCE: 3 T scanners, using T1-weighted 3D fast spoiled gradient-echo sequence, T1-weighted 3D enhanced fast gradient-echo sequence and T1-weighted turbo field echo sequence.

ASSESSMENT: An automated model segmented tumors utilizing DCE-MRI data, followed by a deep learning models (ResNetGN) trained to classify HER2 statuses. Three models were developed to distinguish HER2-zero, HER2-low, and HER2-positive from their respective non-HER2 categories.

STATISTICAL TESTS: Dice similarity coefficient (DSC) was used to evaluate the segmentation performance of the model. Evaluation of the model performances for HER2 statuses involved receiver operating characteristic (ROC) curve analysis and the area under the curve (AUC), accuracy, sensitivity, and specificity. The P-values <0.05 were considered statistically significant.

RESULTS: The automatic segmentation network achieved DSC values of 0.85 to 0.90 compared to the manual segmentation across different sets. The deep learning models using ResNetGN achieved AUCs of 0.782, 0.776, and 0.768 in differentiating HER2-zero from others in the training, internal test, and external test sets, respectively. Similarly, AUCs of 0.820, 0.813, and 0.787 were achieved for HER2-low vs. others, and 0.792, 0.745, and 0.781 for HER2-positive vs. others, respectively.

DATA CONCLUSION: The proposed DCE-MRI-based deep learning system may have the potential to preoperatively distinct HER2 expressions of breast cancers with therapeutic implications.

EVIDENCE LEVEL: 4 TECHNICAL EFFICACY: Stage 3.

PMID:39643475 | DOI:10.1002/jmri.29670

Acad Radiol. 2024 Dec 5:S1076-6332(24)00881-X. doi: 10.1016/j.acra.2024.11.035. Online ahead of print.

ABSTRACT

BACKGROUND/AIM: Contrast-enhanced mammography (CEM) is a relatively novel imaging technique that enables both anatomical and functional breast imaging, with improved diagnostic performance compared to standard 2D mammography. The aim of this study is to systematically review the literature on deep learning (DL) applications for CEM, exploring how these models can further enhance CEM diagnostic potential.

METHODS: This systematic review was reported according to the PRISMA guidelines. We searched for studies published up to April 2024. MEDLINE, Scopus and Google Scholar were used as search databases. Two reviewers independently implemented the search strategy. We included all types of original studies published in English that evaluated DL algorithms for automatic analysis of contrast-enhanced mammography CEM images. The quality of the studies was independently evaluated by two reviewers based on the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) criteria.

RESULTS: Sixteen relevant studies published between 2018 and 2024 were identified. All but one used convolutional neural network models (CNN) models. All studies evaluated DL algorithms for lesion classification, while six studies also assessed lesion detection or segmentation. Segmentation was performed manually in three studies, both manually and automatically in two studies and automatically in ten studies. For lesion classification on retrospective datasets, CNN models reported varied areas under the curve (AUCs) ranging from 0.53 to 0.99. Models incorporating attention mechanism achieved accuracies of 88.1% and 89.1%. Prospective studies reported AUC values of 0.89 and 0.91. Some studies demonstrated that combining DL models with radiomics featured improved classification. Integrating DL algorithms with radiologists' assessments enhanced diagnostic performance.

CONCLUSION: While still at an early research stage, DL can improve CEM diagnostic precision. However, there is a relatively small number of studies evaluating different DL algorithms, and most studies are retrospective. Further prospective testing to assess performance of applications at actual clinical setting is warranted.

PMID:39643464 | DOI:10.1016/j.acra.2024.11.035

Neurosci Biobehav Rev. 2024 Dec 4:105968. doi: 10.1016/j.neubiorev.2024.105968. Online ahead of print.

ABSTRACT

Identifying optimal treatment approaches for schizophrenia is challenging due to varying symptomatology and treatment responses. Artificial intelligence (AI) shows promise in predicting outcomes, prompting this systematic review and meta-analysis to evaluate various AI models' predictive utilities in schizophrenia treatment. A systematic search was conducted, and the risk of bias was evaluated. The pooled sensitivity, specificity, and diagnostic odds ratio with 95% confidence intervals between AI models and the reference standard for response to treatment were assessed. Diagnostic accuracy measures were calculated, and subgroup analysis was performed based on the input data of AI models. Out of the 21 included studies, AI models achieved a pooled sensitivity of 70% and specificity of 76% in predicting schizophrenia treatment response with substantial predictive capacity and a near-to-high level of test accuracy. Subgroup analysis revealed EEG-based models to have the highest sensitivity (89%) and specificity (94%), followed by imaging-based models (76% and 80%, respectively). However, significant heterogeneity was observed across studies in treatment response definitions, participant characteristics, and therapeutic interventions. Despite methodological variations and small sample sizes in some modalities, this study underscores AI's predictive utility in schizophrenia treatment, offering insights for tailored approaches, improving adherence, and reducing relapse risk.

PMID:39643220 | DOI:10.1016/j.neubiorev.2024.105968

Am J Transplant. 2024 Dec 4:S1600-6135(24)00746-9. doi: 10.1016/j.ajt.2024.11.033. Online ahead of print.

ABSTRACT

The COVID-19 pandemic has underscored the importance of vaccines, especially for immunocompromised populations like solid organ transplant (SOT) recipients, who often have weaker immune responses. The purpose of this study was to compare deep learning architectures for predicting SARS-CoV-2 vaccine responses 12 months post-vaccination in this high-risk group. Utilizing data from 303 SOT recipients from a Canadian multicenter cohort, models were developed to forecast anti-receptor binding domain (RBD) antibody levels. The study compared traditional machine learning models-logistic regression, epsilon-support vector regression, random forest regressor, and gradient boosting regressor-and deep learning architectures, including long short-term memory (LSTM), recurrent neural networks, and a novel model, routed LSTM. This new model combines capsule networks with LSTM to reduce the need for large datasets. Demographic, clinical, and transplant-specific data, along with longitudinal antibody measurements, were incorporated into the models. The routed LSTM performed best, achieving a mean square error (MSE) of 0.02±0.02 and a Pearson correlation coefficient (PCC) of 0.79±0.24, outperforming all other models. Key factors influencing vaccine response included age, immunosuppression, breakthrough infection, BMI, sex, and transplant type. These findings suggest that AI could be a valuable tool in tailoring vaccine strategies, improving health outcomes for vulnerable transplant recipients.

PMID:39643006 | DOI:10.1016/j.ajt.2024.11.033

Med Image Anal. 2024 Nov 30;101:103415. doi: 10.1016/j.media.2024.103415. Online ahead of print.

ABSTRACT

Cerebral Microbleeds (CMBs) are chronic deposits of small blood products in the brain tissues, which have explicit relation to various cerebrovascular diseases depending on their anatomical location, including cognitive decline, intracerebral hemorrhage, and cerebral infarction. However, manual detection of CMBs is a time consuming and error-prone process because of their sparse and tiny structural properties. The detection of CMBs is commonly affected by the presence of many CMB mimics that cause a high false-positive rate (FPR), such as calcifications and pial vessels. This paper proposes a novel 3D deep learning framework that not only detects CMBs but also identifies their anatomical location in the brain (i.e., lobar, deep, and infratentorial regions). For the CMBs detection task, we propose a single end-to-end model by leveraging the 3D U-Net as a backbone with Region Proposal Network (RPN). To significantly reduce the false positives within the same single model, we develop a new scheme, containing Feature Fusion Module (FFM) that detects small candidates utilizing contextual information and Hard Sample Prototype Learning (HSPL) that mines CMB mimics and generates additional loss term called concentration loss using Convolutional Prototype Learning (CPL). For the anatomical localization task, we exploit the 3D U-Net segmentation network to segment anatomical structures of the brain. This task not only identifies to which region the CMBs belong but also eliminates some false positives from the detection task by leveraging anatomical information. We utilize Susceptibility-Weighted Imaging (SWI) and phase images as 3D input to efficiently capture 3D information. The results show that the proposed RPN that utilizes the FFM and HSPL outperforms the baseline RPN and achieves a sensitivity of 94.66 % vs. 93.33 % and an average number of false positives per subject (FPavg) of 0.86 vs. 14.73. Furthermore, the anatomical localization task enhances the detection performance by reducing the FPavg to 0.56 while maintaining the sensitivity of 94.66 %.

PMID:39642804 | DOI:10.1016/j.media.2024.103415

J Hazard Mater. 2024 Nov 28;484:136558. doi: 10.1016/j.jhazmat.2024.136558. Online ahead of print.

ABSTRACT

Monitoring radioactive cesium ions (Cs+) in seawater is vital for environmental safety but remains challenging due to limitations in the accessibility, stability, and selectivity of traditional methods. This study presents an innovative approach that combines electrochemical voltammetry using nickel hexacyanoferrate (NiHCF) thin-film electrode with machine learning (ML) to enable accurate and portable detection of Cs+. Optimizing the fabrication of NiHCF thin-film electrodes enabled the development of a robust sensor that generates cyclic voltammograms (CVs) sensitive to Cs⁺ concentrations as low as 1 ppb in synthetic seawater and 10 ppb in real seawater, with subtle changes in CV patterns caused by trace Cs⁺ effectively identified and analyzed using ML. Using 2D convolutional neural networks (CNNs), we classified Cs+ concentrations across eight logarithmic classes (0 - 106 ppb) with 100 % accuracy and an F1-score of 1 in synthetic seawater datasets, outperforming the 1D CNN and deep neural networks. Validation using real seawater datasets confirmed the applicability of our model, achieving high performance. Moreover, gradient-weighted class activation mapping (Grad-CAM) identified critical CV regions that were overlooked during manual inspection, validating model reliability. This integrated method offers sensitive and practical solutions for monitoring Cs+ in seawater, helping to prevent its accumulation in ecosystems.

PMID:39642734 | DOI:10.1016/j.jhazmat.2024.136558

Comput Biol Med. 2024 Dec 5;185:109459. doi: 10.1016/j.compbiomed.2024.109459. Online ahead of print.

ABSTRACT

The liver is one of the vital organs in the body. Precise liver segmentation in medical images is essential for liver disease treatment. The deep learning-based liver segmentation process faces several challenges. This research aims to analyze the challenges of liver segmentation in prior studies and identify the modifications made to network models and other enhancements implemented by researchers to tackle each challenge. In total, 88 articles from Scopus and ScienceDirect databases published between January 2016 and January 2022 have been studied. The liver segmentation challenges are classified into five main categories, each containing some subcategories. For each challenge, the proposed technique to overcome the challenge is investigated. The provided report details the authors, publication years, dataset types, imaging technologies, and evaluation metrics of all references for comparison. Additionally, a summary table outlines the challenges and solutions.

PMID:39642700 | DOI:10.1016/j.compbiomed.2024.109459

Med Phys. 2024 Dec 6. doi: 10.1002/mp.17554. Online ahead of print.

ABSTRACT

BACKGROUND: Surgical planning for orthognathic procedures demands swift and accurate biomechanical modeling of facial soft tissues. Efficient simulations are vital in the clinical pipeline, as surgeons may iterate through multiple plans. Biomechanical simulations typically use the finite element method (FEM). Prior works divide FEM simulations into increments to enhance convergence and accuracy. However, this practice elongates simulation time, thereby impeding clinical integration. To accelerate simulations, deep learning (DL) models have been explored. Yet, previous efforts either perform simulations in a single step or neglect the temporal aspects in incremental simulations.

PURPOSE: This study investigates the use of spatiotemporal incremental modeling for biomechanics simulations of facial soft tissue.

METHODS: We implement the method using a graph neural network. Our method synergizes spatial features with temporal aggregation using DL networks trained on incremental FEM simulations from 17 subjects that underwent orthognathic surgery.

RESULTS: Our proposed spatiotemporal incremental method achieved a mean accuracy of 0.37 mm with a mean computation time of 1.52 s. In comparison, a spatial-only incremental method yielded a mean accuracy of 0.44 mm and a mean computation time of 1.60 s, while a spatial-only single-step method yielded a mean accuracy of 0.41 mm and a mean computation time of 0.05 s.

CONCLUSIONS: Statistical analysis demonstrated that the spatiotemporal incremental method reduced mean errors compared to the spatial-only incremental method, emphasizing the importance of incorporating temporal information in incremental simulations. Overall, we successfully implemented spatiotemporal incremental learning tailored to simulate soft tissue deformation while substantially reducing simulation time compared to FEM.

PMID:39642013 | DOI:10.1002/mp.17554

Med Phys. 2024 Dec 6. doi: 10.1002/mp.17562. Online ahead of print.

ABSTRACT

BACKGROUND: Magnetic resonance imaging (MRI) is a crucial technique for both scientific research and clinical diagnosis. However, noise generated during MR data acquisition degrades image quality, particularly in hyperpolarized (HP) gas MRI. While deep learning (DL) methods have shown promise for MR image denoising, most of them fail to adequately utilize the long-range information which is important to improve denoising performance. Furthermore, the sample size of paired noisy and noise-free MR images also limits denoising performance.

PURPOSE: To develop an effective DL method that enhances denoising performance and reduces the requirement of paired MR images by utilizing the long-range information and pretraining.

METHODS: In this work, a hybrid Transformer-convolutional neural network (CNN) network (HTC-net) and a self-supervised pretraining strategy are proposed, which effectively enhance the denoising performance. In HTC-net, a CNN branch is exploited to extract the local features. Then a Transformer-CNN branch with two parallel encoders is designed to capture the long-range information. Within this branch, a residual fusion block (RFB) with a residual feature processing module and a feature fusion module is proposed to aggregate features at different resolutions extracted by two parallel encoders. After that, HTC-net exploits the comprehensive features from the CNN branch and the Transformer-CNN branch to accurately predict noise-free MR images through a reconstruction module. To further enhance the performance on limited MRI datasets, a self-supervised pretraining strategy is proposed. This strategy employs self-supervised denoising to equip the HTC-net with denoising capabilities during pretraining, and then the pre-trained parameters are transferred to facilitate subsequent supervised training.

RESULTS: Experimental results on the pulmonary HP 129Xe MRI dataset (1059 images) and IXI dataset (5000 images) all demonstrate the proposed method outperforms the state-of-the-art methods, exhibiting superior preservation of edges and structures. Quantitatively, on the pulmonary HP 129Xe MRI dataset, the proposed method outperforms the state-of-the-art methods by 0.254-0.597 dB in PSNR and 0.007-0.013 in SSIM. On the IXI dataset, the proposed method outperforms the state-of-the-art methods by 0.3-0.927 dB in PSNR and 0.003-0.016 in SSIM.

CONCLUSIONS: The proposed method can effectively enhance the quality of MR images, which helps improve the diagnosis accuracy in clinical.

PMID:39641989 | DOI:10.1002/mp.17562

Magn Reson Med. 2024 Dec 6. doi: 10.1002/mrm.30401. Online ahead of print.

ABSTRACT

PURPOSE: To develop a deep learning-based method for robust and rapid estimation of the fatty acid composition (FAC) in mammary adipose tissue.

METHODS: A physics-based unsupervised deep learning network for estimation of fatty acid composition-network (FAC-Net) is proposed to estimate the number of double bonds and number of methylene-interrupted double bonds from multi-echo bipolar gradient-echo data, which are subsequently converted to saturated, mono-unsaturated, and poly-unsaturated fatty acids. The loss function was based on a 10 fat peak signal model. The proposed network was tested with a phantom containing eight oils with different FAC and on post-menopausal women scanned using a whole-body 3T MRI system between February 2022 and January 2024. The post-menopausal women included a control group (n = 8) with average risk for breast cancer and a cancer group (n = 7) with biopsy-proven breast cancer.

RESULTS: The FAC values of eight oils in the phantom showed strong correlations between the measured and reference values (R2 > 0.9 except chain length). The FAC values measured from scan and rescan data of the control group showed no significant difference between the two scans. The FAC measurements of the cancer group conducted before contrast and after contrast showed a significant difference in saturated fatty acid and mono-unsaturated fatty acid. The cancer group has higher saturated fatty acid than the control group, although not statistically significant.

CONCLUSION: The results in this study suggest that the proposed FAC-Net can be used to measure the FAC of mammary adipose tissue from gradient-echo MRI data of the breast.

PMID:39641987 | DOI:10.1002/mrm.30401

Bull Math Biol. 2024 Dec 6;87(1):12. doi: 10.1007/s11538-024-01379-w.

ABSTRACT

The [ P S I + ] prion phenotype in yeast manifests as a white, pink, or red color pigment. Experimental manipulations destabilize prion phenotypes, and allow colonies to exhibit [ p s i - ] (red) sectored phenotypes within otherwise completely white colonies. Further investigation of the size and frequency of sectors that emerge as a result of experimental manipulation is capable of providing critical information on mechanisms of prion curing, but we lack a way to reliably extract this information. Images of experimental colonies exhibiting sectored phenotypes offer an abundance of data to help uncover molecular mechanisms of sectoring, yet the structure of sectored colonies is ignored in traditional biological pipelines. In this study, we present [PSI]-CIC, the first computational pipeline designed to identify and characterize features of sectored yeast colonies. To overcome the barrier of a lack of manually annotated data of colonies, we develop a neural network architecture that we train on synthetic images of colonies and apply to real images of [ P S I + ] , [ p s i - ] , and sectored colonies. In hand-annotated experimental images, our pipeline correctly predicts the state of approximately 95% of colonies detected and frequency of sectors in approximately 89.5% of colonies detected. The scope of our pipeline could be extended to categorizing colonies grown under different experimental conditions, allowing for more meaningful and detailed comparisons between experiments. Our approach streamlines the analysis of sectored yeast colonies providing a rich set of quantitative metrics and provides insight into mechanisms driving the curing of prion phenotypes.

PMID:39641894 | DOI:10.1007/s11538-024-01379-w

JACC Cardiovasc Imaging. 2024 Nov 22:S1936-878X(24)00417-0. doi: 10.1016/j.jcmg.2024.09.010. Online ahead of print.

ABSTRACT

BACKGROUND: Cardiac magnetic resonance (CMR) imaging is an important diagnostic tool for diagnosis of cardiac amyloidosis (CA). However, discrimination of CA from other etiologies of myocardial disease can be challenging.

OBJECTIVES: The aim of this study was to develop and rigorously validate a deep learning (DL) algorithm to aid in the discrimination of CA using cine and late gadolinium enhancement CMR imaging.

METHODS: A DL model using a retrospective cohort of 807 patients who were referred for CMR for suspicion of infiltrative disease or hypertrophic cardiomyopathy (HCM) was developed. Confirmed definitive diagnosis was as follows: 252 patients with CA, 290 patients with HCM, and 265 with neither CA or HCM (other). This cohort was split 70/30 into training and test sets. A vision transformer (ViT) model was trained primarily to identify CA. The model was validated in an external cohort of 157 patients also referred for CMR for suspicion of infiltrative disease or HCM (51 CA, 49 HCM, 57 other).

RESULTS: The ViT model achieved a diagnostic accuracy (84.1%) and an area under the curve of 0.954 in the internal testing data set. The ViT model further demonstrated an accuracy of 82.8% and an area under the curve of 0.957 in the external testing set. The ViT model achieved an accuracy of 90% (n = 55 of 61), among studies with clinical reports with moderate/high confidence diagnosis of CA, and 61.1% (n = 22 of 36) among studies with reported uncertain, missing, or incorrect diagnosis of CA in the internal cohort. DL accuracy of this cohort increased to 79.1% when studies with poor image quality, dual pathologies, or ambiguity of clinically significant CA diagnosis were removed.

CONCLUSIONS: A ViT model using only cine and late gadolinium enhancement CMR images can achieve high accuracy in differentiating CA from other underlying etiologies of suspected cardiomyopathy, especially in cases when reported human diagnostic confidence was uncertain in both a large single state health system and in an external CA cohort.

PMID:39641685 | DOI:10.1016/j.jcmg.2024.09.010

Heliyon. 2024 Oct 3;10(22):e38726. doi: 10.1016/j.heliyon.2024.e38726. eCollection 2024 Nov 30.

ABSTRACT

BACKGROUND AND OBJECTIVE: The multi-targets and multi-components of Traditional Chinese medicine (TCM) coincide with the complex pathogenesis of depression. Zhi-Zi-Hou-Pu Decoction (ZZHPD) has been approved in clinical medication with good antidepression effects for centuries, while the mechanisms under the iceberg haven't been addressed systematically. This study explored its inner active ingredients - potent pharmacological mechanism - DDI to explore more comprehensively and deeply understanding of the complicated TCM in treatment.

METHODS: This research utilized network pharmacology combined with molecular docking to identify pharmacological targets and molecular interactions between ZZHPD and depression. Verification of major active compounds was conducted through UPLC-Q-TOF-MS/MS and assays on LPS-induced neuroblastoma cells. Additionally, the DDIMDL model, a deep learning-based approach, was used to predict DDIs, focusing on serum concentration, metabolism, effectiveness, and adverse reactions.

RESULTS: The antidepressant mechanisms of ZZHPD involve the serotonergic synapse, neuroactive ligand-receptor interaction, and dopaminergic synapse signaling pathways. Eighteen active compounds were identified, with honokiol and eriocitrin significantly modulating neuronal inflammation and promoting differentiation of neuroimmune cells through genes like COMT, PI3KCA, PTPN11, and MAPK1. DDI predictions indicated that eriocitrin's serum concentration increases when combined with hesperidin, while hesperetin's metabolism decreases with certain flavonoids. These findings provide crucial insights into the nervous system's effectiveness and potential cardiovascular or nervous system adverse reactions from core compound combinations.

CONCLUSIONS: This study provides insights into the TCM interpretation, drug compatibility or combined medication for further clinical application or potential drug pairs with a cost-effective method of integrated network pharmacology and deep learning.

PMID:39641032 | PMC:PMC11617927 | DOI:10.1016/j.heliyon.2024.e38726