Heliyon. 2024 Nov 14;10(22):e40345. doi: 10.1016/j.heliyon.2024.e40345. eCollection 2024 Nov 30.

ABSTRACT

Cyclin Dependent Kinase 1 (CDK1) plays a crucial role in cell cycle regulation, and dysregulation of its activity has been implicated in various cancers. Although several CDK1 inhibitors are currently in clinical trials, none have yet been approved for therapeutic use. This research utilized deep learning techniques, specifically Recurrent Neural Networks with Long Short-Term Memory (LSTM), to generate potential CDK1 inhibitors. Molecular docking, evaluation of molecular properties, and molecular dynamics simulations were conducted to identify the most promising candidates. The results showed that the generated ligands exhibited substantial improvements in target affinity and drug-likeness. Molecular docking results showed that the generated ligands had an average binding affinity of -10.65 ± 0.877 kcal/mol towards CDK1. The Quantitative Estimate of Drug-likeness (QED) values for the generated ligands averaged 0.733 ± 0.10, significantly higher than the 0.547 ± 0.15 observed for known CDK1 inhibitors (p < 0.001). Molecular dynamics simulations further confirmed the stability and favorable interactions of the selected ligands with the CDK1 complex. The identification of novel CDK1 inhibitors with improved binding affinities and drug-likeness properties could potentially fill the gap in the ongoing development of CDK inhibitors. However, it is imperative to note that extensive experimental validation is required prior to advancing these generated ligands to subsequent stages of drug development.

PMID:39748968 | PMC:PMC11693894 | DOI:10.1016/j.heliyon.2024.e40345

IEEE Access. 2024;12:191219-191230. doi: 10.1109/access.2024.3516572. Epub 2024 Dec 12.

ABSTRACT

Given telemetry datasets (e.g., GPS location, speed, direction, distance.), the Older Driver Classification (ODC) problem identifies two groups of drivers: normal and abnormal. The ODC problem is essential in many societal applications, including road safety, insurance risk assessment, and targeted interventions for elderly drivers with dementia or Mild Cognitive Impairment (MCI). The problem is challenging because of the volume and heterogeneity of temporally-detailed vehicle datasets. This paper proposes a novel spatial deep-learning approach that leverages Grid-Index based data augmentation to enhance the detection of abnormal driving behaviors. Through extensive experiments and a real-world case study, the proposed approach consistently identifies abnormal drivers with high accuracy. The findings demonstrate the potential of grid-based methods to improve telematics-based driving behavior analysis significantly. This approach offers valuable implications for enhancing road safety measures, optimizing insurance risk assessments, and developing targeted interventions for at-risk drivers.

PMID:39748855 | PMC:PMC11694628 | DOI:10.1109/access.2024.3516572

Front Plant Sci. 2024 Dec 19;15:1512632. doi: 10.3389/fpls.2024.1512632. eCollection 2024.

ABSTRACT

Molecular-based detection of pathogens from potato tubers hold promise, but the initial sample extraction process is labor-intensive. Developing a robotic tuber sampling system, equipped with a fast and precise machine vision technique to identify optimal sampling locations on a potato tuber, offers a viable solution. However, detecting sampling locations such as eyes and stolon scar is challenging due to variability in their appearance, size, and shape, along with soil adhering to the tubers. In this study, we addressed these challenges by evaluating various deep-learning-based object detectors, encompassing You Look Only Once (YOLO) variants of YOLOv5, YOLOv6, YOLOv7, YOLOv8, YOLOv9, YOLOv10, and YOLO11, for detecting eyes and stolon scars across a range of diverse potato cultivars. A robust image dataset obtained from tubers of five potato cultivars (three russet skinned, a red skinned, and a purple skinned) was developed as a benchmark for detection of these sampling locations. The mean average precision at an intersection over union threshold of 0.5 (mAP@0.5) ranged from 0.832 and 0.854 with YOLOv5n to 0.903 and 0.914 with YOLOv10l. Among all the tested models, YOLOv10m showed the optimal trade-off between detection accuracy (mAP@0.5 of 0.911) and inference time (92 ms), along with satisfactory generalization performance when cross-validated among the cultivars used in this study. The model benchmarking and inferences of this study provide insights for advancing the development of a robotic potato tuber sampling device.

PMID:39748820 | PMC:PMC11693691 | DOI:10.3389/fpls.2024.1512632

Mayo Clin Proc Digit Health. 2024 Dec;2(4):548-558. doi: 10.1016/j.mcpdig.2024.08.005. Epub 2024 Aug 26.

ABSTRACT

OBJECTIVE: To report the development and performance of 2 distinct deep learning models trained exclusively on retinal color fundus photographs to classify Alzheimer disease (AD).

PATIENTS AND METHODS: Two independent datasets (UK Biobank and our tertiary academic institution) of good-quality retinal photographs derived from patients with AD and controls were used to build 2 deep learning models, between April 1, 2021, and January 30, 2024. ADVAS is a U-Net-based architecture that uses retinal vessel segmentation. ADRET is a bidirectional encoder representations from transformers style self-supervised learning convolutional neural network pretrained on a large data set of retinal color photographs from UK Biobank. The models' performance to distinguish AD from non-AD was determined using mean accuracy, sensitivity, specificity, and receiving operating curves. The generated attention heatmaps were analyzed for distinctive features.

RESULTS: The self-supervised ADRET model had superior accuracy when compared with ADVAS, in both UK Biobank (98.27% vs 77.20%; P<.001) and our institutional testing data sets (98.90% vs 94.17%; P=.04). No major differences were noted between the original and binary vessel segmentation and between both eyes vs single-eye models. Attention heatmaps obtained from patients with AD highlighted regions surrounding small vascular branches as areas of highest relevance to the model decision making.

CONCLUSION: A bidirectional encoder representations from transformers style self-supervised convolutional neural network pretrained on a large data set of retinal color photographs alone can screen symptomatic AD with high accuracy, better than U-Net-pretrained models. To be translated in clinical practice, this methodology requires further validation in larger and diverse populations and integrated techniques to harmonize fundus photographs and attenuate the imaging-associated noise.

PMID:39748801 | PMC:PMC11695061 | DOI:10.1016/j.mcpdig.2024.08.005

Pathologica. 2024 Dec;116(6):390-403. doi: 10.32074/1591-951X-1069.

ABSTRACT

The search for reliable prognostic markers in oral squamous cell carcinoma (OSCC) remains a critical need. Tumor-infiltrating lymphocytes (TILs), particularly T lymphocytes, play a pivotal role in the immune response against tumors and are strongly correlated with favorable prognoses. Computational pathology has proven highly effective for histopathological image analysis, automating tasks such as cell detection, classification, and segmentation.

In the present study, we developed a StarDist-based model to automatically detect T lymphocytes in hematoxylin and eosin (H&E)-stained whole-slide images (WSIs) of OSCC, bypassing the need for traditional immunohistochemistry (IHC). Using QuPath, we generated training datasets from annotated slides, employing IHC as the ground truth. Our model was validated on Cancer Genome Atlas-derived OSCC images, and survival analyses demonstrated that higher TIL densities correlated with improved patient outcomes.

This work introduces an efficient, AI-powered workflow for automated immune profiling in OSCC, offering a reproducible and scalable approach for diagnostic and prognostic applications.

PMID:39748724 | DOI:10.32074/1591-951X-1069

Microsc Res Tech. 2025 Jan 2. doi: 10.1002/jemt.24788. Online ahead of print.

ABSTRACT

Microscopic imaging aids disease diagnosis by describing quantitative cell morphology and tissue size. However, the high spatial resolution of these images poses significant challenges for manual quantitative evaluation. This project proposes using computer-aided analysis methods to address these challenges, enabling rapid and precise clinical diagnosis, course analysis, and prognostic prediction. This research introduces advanced deep learning frameworks such as squeeze-and-excitation and dilated dense convolution blocks to tackle the complexities of quantifying small and intricate breast cancer tissues and meeting the real-time requirements of pathological image analysis. Our proposed framework integrates a dense convolutional network (DenseNet) with an attention mechanism, enhancing the capability for rapid and accurate clinical assessments. These multi-classification models facilitate the precise prediction and segmentation of breast lesions in microscopic images by leveraging lightweight multi-scale feature extraction, dynamic region attention, sub-region classification, and regional regularization loss functions. This research will employ transfer learning paradigms and data enhancement methods to enhance the models' learning further and prevent overfitting. We propose the fine-tuning employing pre-trained architectures such as VGGNet-19, ResNet152V2, EfficientNetV2-B1, and DenseNet-121, modifying the final pooling layer in each model's last block with an SPP layer and associated BN layer. The study uses labeled and unlabeled data for tissue microscopic image analysis, enhancing models' robust features and classification abilities. This method reduces the costs and time associated with traditional methods, alleviating the burden of data labeling in computational pathology. The goal is to provide a sophisticated, efficient quantitative pathological image analysis solution, improving clinical outcomes and advancing the computational field. The model, trained, validated, and tested on a microscope breast image dataset, achieved recognition accuracy of 99.6% for benign and malignant secondary classification and 99.4% for eight breast subtypes classification. Our proposed approach demonstrates substantial improvement compared to existing methods, which generally report lower accuracies for breast subtype classification ranging between 85% and 94%. This high level of accuracy underscores the potential of our approach to provide reliable diagnostic support, enhancing precision in clinical decision-making.

PMID:39748498 | DOI:10.1002/jemt.24788

Cardiovasc Diabetol. 2025 Jan 2;24(1):3. doi: 10.1186/s12933-024-02564-w.

ABSTRACT

BACKGROUND: Prior studies have demonstrated an association between retinal vascular features and cardiovascular disease (CVD), however most studies have only evaluated a few simple parameters at a time. Our aim was to determine whether a deep-learning artificial intelligence (AI) model could be used to predict CVD outcomes from routinely obtained diabetic retinal screening photographs and to compare its performance to a traditional clinical CVD risk score.

METHODS: We included 6127 individuals with type 2 diabetes without myocardial infarction or stroke prior to study entry. The cohort was divided into training (70%), validation (10%) and testing (20%) cohorts. Clinical 10-year CVD risk was calculated using the pooled cohort equation (PCE) risk score. A polygenic risk score (PRS) for coronary heart disease was also obtained. Retinal images were analysed using an EfficientNet-B2 network to predict 10-year CVD risk. The primary outcome was time to first major adverse CV event (MACE) including CV death, myocardial infarction or stroke.

RESULTS: 1241 individuals were included in the test cohort (mean PCE 10-year CVD risk 35%). There was a strong correlation between retinal predicted CVD risk and the PCE risk score (r = 0.66) but not the polygenic risk score (r = 0.05). There were 288 MACE events. Higher retina-predicted risk was significantly associated with increased 10-year risk of MACE (HR 1.05 per 1% increase; 95% CI 1.04-1.06, p < 0.001) and remained so after adjustment for the PCE and polygenic risk score (HR 1.03; 95% CI 1.02-1.04, p < 0.001). The retinal risk score had similar performance to the PCE (both AUC 0.697) and when combined with the PCE and polygenic risk score had significantly improved performance compared to the PCE alone (AUC 0.728). An increase in retinal-predicted risk within 3 years was associated with subsequent increased MACE likelihood.

CONCLUSIONS: A deep-learning AI model could accurately predict MACE from routine retinal screening photographs with a comparable performance to traditional clinical risk assessment in a diabetic cohort. Combining the AI-derived retinal risk prediction with a coronary heart disease polygenic risk score improved risk prediction. AI retinal assessment might allow a one-stop CVD risk assessment at routine retinal screening.

PMID:39748380 | DOI:10.1186/s12933-024-02564-w

Sci Rep. 2025 Jan 2;15(1):230. doi: 10.1038/s41598-024-81862-5.

ABSTRACT

MiRNAs and lncRNAs are two essential noncoding RNAs. Predicting associations between noncoding RNAs and diseases can significantly improve the accuracy of early diagnosis.With the continuous breakthroughs in artificial intelligence, researchers increasingly use deep learning methods to predict associations. Nevertheless, most existing methods face two major issues: low prediction accuracy and the limitation of only being able to predict a single type of noncoding RNA-disease association. To address these challenges, this paper proposes a method called K-Means and multigraph Contrastive Learning for predicting associations among miRNAs, lncRNAs, and diseases (K-MGCMLD). The K-MGCMLD model is divided into four main steps. The first step is the construction of a heterogeneous graph. The second step involves down sampling using the K-means clustering algorithm to balance the positive and negative samples. The third step is to use an encoder with a Graph Convolutional Network (GCN) architecture to extract embedding vectors. Multigraph contrastive learning, including both local and global graph contrastive learning, is used to help the embedding vectors better capture the latent topological features of the graph. The fourth step involves feature reconstruction using the balanced positive and negative samples and the embedding vectors fed into an XGBoost classifier for multi-association classification prediction. Experimental results have shown that AUC value for miRNA-disease association is 0.9542, lncRNA-disease association is 0.9603, and lncRNA-miRNA association is 0.9687. Additionally, this study has conducted case analyses using K-MGCMLD, which has validated the associations of all the top 30 miRNAs predicted to be associated with lung cancer and Alzheimer's diseases.

PMID:39747154 | DOI:10.1038/s41598-024-81862-5

Nat Commun. 2025 Jan 2;16(1):136. doi: 10.1038/s41467-024-54970-z.

ABSTRACT

Microsatellite instability (MSI) is a critical phenotype of cancer genomes and an FDA-recognized biomarker that can guide treatment with immune checkpoint inhibitors. Previous work has demonstrated that next-generation sequencing data can be used to identify samples with MSI-high phenotype. However, low tumor purity, as frequently observed in routine clinical samples, poses a challenge to the sensitivity of existing algorithms. To overcome this critical issue, we developed MiMSI, an MSI classifier based on deep neural networks and trained using a dataset that included low tumor purity MSI cases in a multiple instance learning framework. On a challenging yet representative set of cases, MiMSI showed higher sensitivity (0.895) and auROC (0.971) than MSISensor (sensitivity: 0.67; auROC: 0.907), an open-source software previously validated for clinical use at our institution using MSK-IMPACT large panel targeted NGS data. In a separate, prospective cohort, MiMSI confirmed that it outperforms MSISensor in low purity cases (P = 8.244e-07).

PMID:39746944 | DOI:10.1038/s41467-024-54970-z

Mol Cell Proteomics. 2024 Dec 31:100901. doi: 10.1016/j.mcpro.2024.100901. Online ahead of print.

ABSTRACT

Animal venoms, distinguished by their unique structural features and potent bioactivities, represent a vast and relatively untapped reservoir of therapeutic molecules. However, limitations associated with comprehensively constructing and expressing highly complex venom and venom-like molecule libraries have precluded their therapeutic evaluation via high throughput screening. Here, we developed an innovative computational approach to design a highly diverse library of animal venoms and "metavenoms". We employed programmable M13 hyperphage display to preserve critical disulfide-bonded structures for highly parallelized single-round biopanning with quantitation via high-throughput DNA sequencing. Our approach led to the discovery of Kunitz type domain containing proteins that target the human itch receptor Mas-related G protein-coupled receptor X4 (MRGPRX4), which plays a crucial role in itch perception. Deep learning-based structural homology mining identified two endogenous human homologs, tissue factor pathway inhibitor (TFPI) and serine peptidase inhibitor, Kunitz type 2 (SPINT2), which exhibit agonist-dependent potentiation of MRGPRX4. Highly multiplexed screening of animal venoms and metavenoms is therefore a promising approach to uncover new drug candidates.

PMID:39746545 | DOI:10.1016/j.mcpro.2024.100901

J Biomed Inform. 2024 Dec 31:104745. doi: 10.1016/j.jbi.2024.104745. Online ahead of print.

ABSTRACT

OBJECTIVE: Recognizing glomerular lesions is essential in diagnosing chronic kidney disease. However, deep learning faces challenges due to the lesion heterogeneity, superposition, progression, and tissue incompleteness, leading to uncertainty in model predictions. Therefore, it is crucial to analyze pathology-related predictive uncertainty in glomerular lesion recognition and unveil its relationship with pathological properties and its impact on model performance.

METHODS: This paper presents a novel framework for pathology-related predictive uncertainty analysis towards glomerular lesion recognition, including prototype learning based predictive uncertainty estimation, pathology-characterized correlation analysis and weight-redistributed prediction rectification. The prototype learning based predictive uncertainty estimation includes deep prototyping, affinity embedding, and multi-dimensional uncertainty fusion. The pathology-characterized correlation analysis is the first to use expert-based and learning- based approach to construct the pathology-related characterization of lesions and tissues. The weight-redistributed prediction rectification module performs reweighting- based lesion recognition.

RESULTS: To validate the performance, extensive experiments were conducted. Based on the Spearman and Pearson correlation analysis the proposed framework enables more efficient correlation analysis, and strong correlation with pathology-related characterization can be achieved (c index > 0.6 and p < 0.01). Furthermore, the prediction rectification module demonstrated improved lesion recognition performance across most metrics, with enhancements of up to 6.36 %.

CONCLUSION: The proposed predictive uncertainty analysis in glomerular lesion recognition offers a valuable approach for assessing computational pathology's predictive uncertainty from a pathology-related perspective.

SIGNIFICANCE: The paper provides a solution for pathology-related predictive uncertainty estimation in algorithm development and clinical practice.

PMID:39746430 | DOI:10.1016/j.jbi.2024.104745

Eur Radiol. 2025 Jan 2. doi: 10.1007/s00330-024-11256-8. Online ahead of print.

ABSTRACT

OBJECTIVES: Incidentally detected pulmonary nodules present a challenge in clinical routine with demand for reliable support systems for risk classification. We aimed to evaluate the performance of the lung-cancer-prediction-convolutional-neural-network (LCP-CNN), a deep learning-based approach, in comparison to multiparametric statistical methods (Brock model and Lung-RADS®) for risk classification of nodules in cohorts with different risk profiles and underlying pulmonary diseases.

MATERIALS AND METHODS: Retrospective analysis was conducted on non-contrast and contrast-enhanced CT scans containing pulmonary nodules measuring 5-30 mm. Ground truth was defined by histology or follow-up stability. The final analysis was performed on 297 patients with 422 eligible nodules, of which 105 nodules were malignant. Classification performance of the LCP-CNN, Brock model, and Lung-RADS® was evaluated in terms of diagnostic accuracy measurements including ROC-analysis for different subcohorts (total, screening, emphysema, and interstitial lung disease).

RESULTS: LCP-CNN demonstrated superior performance compared to the Brock model in total and screening cohorts (AUC 0.92 (95% CI: 0.89-0.94) and 0.93 (95% CI: 0.89-0.96)). Superior sensitivity of LCP-CNN was demonstrated compared to the Brock model and Lung-RADS® in total, screening, and emphysema cohorts for a risk threshold of 5%. Superior sensitivity of LCP-CNN was also shown across all disease groups compared to the Brock model at a threshold of 65%, compared to Lung-RADS® sensitivity was better or equal. No significant differences in the performance of LCP-CNN were found between subcohorts.

CONCLUSION: This study offers further evidence of the potential to integrate deep learning-based decision support systems into pulmonary nodule classification workflows, irrespective of the individual patient risk profile and underlying pulmonary disease.

KEY POINTS: Question Is a deep-learning approach (LCP-CNN) superior to multiparametric models (Brock model, Lung-RADS®) in classifying pulmonary nodule risk across varied patient profiles? Findings LCP-CNN shows superior performance in risk classification of pulmonary nodules compared to multiparametric models with no significant impact on risk profiles and structural pulmonary diseases. Clinical relevance LCP-CNN offers efficiency and accuracy, addressing limitations of traditional models, such as variations in manual measurements or lack of patient data, while producing robust results. Such approaches may therefore impact clinical work by complementing or even replacing current approaches.

PMID:39747589 | DOI:10.1007/s00330-024-11256-8

Sci Rep. 2025 Jan 2;15(1):113. doi: 10.1038/s41598-024-84071-2.

ABSTRACT

Symbolic anchoring is an important topic in robotics, as it enables robots to obtain symbolic knowledge from the perceptual information acquired through their sensors and maintain the link between that knowledge and the sensory data. In cognitive-based robots, this process of transforming sub-symbolic data generated by sensors to obtain and maintain symbolic knowledge is still an open problem. To address this issue, this paper presents SAILOR, a framework for symbolic anchoring integrated into ROS 2. SAILOR aims to maintain the link between symbolic data and perceptual data in real robots over time. It provides a semantic world modeling approach using two deep learning-based sub-symbolic robotic skills: object recognition and matching function. The object recognition skill allows the robot to recognize and identify objects in its environment, while the matching function enables the robot to decide if new perceptual data corresponds to existing symbolic data. This paper describes the proposed method and the development of the framework, as well as its integration in MERLIN2 (a hybrid cognitive architecture fully functional in robots running ROS 2) and the validation of SAILOR using public datasets and a real-world scenario.

PMID:39747469 | DOI:10.1038/s41598-024-84071-2

Sci Rep. 2025 Jan 2;15(1):54. doi: 10.1038/s41598-024-83924-0.

ABSTRACT

The hypotensive side effects caused by drugs during their use have been a vexing issue. Recent studies have found that deep learning can effectively predict the biological activity of compounds by mining patterns and rules in the data, providing a potential solution for identifying drug side effects. In this study, we established a deep learning-based predictive model, utilizing a data set comprised of compounds known to either elevate or lower blood pressure. Subsequently, the trained model was used to predict the blood pressure-lowering effects of 26,000 compounds. Based on the predicted results, we randomly selected 50 molecules for validation and compared them with literature reports. The results showed that the predictions for 30 molecules were consistent with literature reports, with known antihypertensive drugs such as reserpine, guanethidine, and mecamylamine ranking at the top. We further selected 10 of these molecules and 3 related protein targets for molecular docking, and the docking results indirectly confirmed the model's accuracy. Ultimately, we discovered and validated that salaprinol significantly inhibits ACE1 activity and lowers canine blood pressure. In summary, we have established a highly accurate activity prediction model and confirmed its accuracy in predicting potential blood pressure-lowering compounds, which is expected to help patients avoid hypotensive side effects during clinical medication and also provide significant assistance in the discovery of antihypertensive drugs.

PMID:39747442 | DOI:10.1038/s41598-024-83924-0

Sci Rep. 2025 Jan 2;15(1):207. doi: 10.1038/s41598-024-84673-w.

ABSTRACT

Insider threats pose a significant challenge to IT security, particularly with the rise of generative AI technologies, which can create convincing fake user profiles and mimic legitimate behaviors. Traditional intrusion detection systems struggle to differentiate between real and AI-generated activities, creating vulnerabilities in detecting malicious insiders. To address this challenge, this paper introduces a novel Deep Synthesis Insider Intrusion Detection (DS-IID) model. The model employs deep feature synthesis to automatically generate detailed user profiles from event data and utilizes binary deep learning for accurate threat identification. The DS-IID model addresses three key issues: it (i) detects malicious insiders using supervised learning, (ii) evaluates the effectiveness of generative algorithms in replicating real user profiles, and (iii) distinguishes between real and synthetic abnormal user profiles. To handle imbalanced data, the model uses on-the-fly weighted random sampling. Tested on the CERT insider threat dataset, the DS-IID achieved 97% accuracy and an AUC of 0.99. Moreover, the model demonstrates strong performance in differentiating real from AI-generated (synthetic) threats, achieving over 99% accuracy on optimally generated data. While primarily evaluated on synthetic datasets, the high accuracy of the DS-IID model suggests its potential as a valuable tool for real-world cybersecurity applications.

PMID:39747424 | DOI:10.1038/s41598-024-84673-w

Sci Rep. 2025 Jan 2;15(1):457. doi: 10.1038/s41598-024-84748-8.

ABSTRACT

In modern agriculture, the proliferation of weeds in cotton fields poses a significant threat to the healthy growth and yield of crops. Therefore, efficient detection and control of cotton field weeds are of paramount importance. In recent years, deep learning models have shown great potential in the detection of cotton field weeds, achieving high-precision weed recognition. However, existing deep learning models, despite their high accuracy, often have complex computations and high resource consumption, making them difficult to apply in practical scenarios. To address this issue, developing efficient and lightweight detection methods for weed recognition in cotton fields is crucial for effective weed control. This study proposes the YOLO-Weed Nano algorithm based on the improved YOLOv8n model. First, the Depthwise Separable Convolution (DSC) structure is used to improve the HGNetV2 network, creating the DS_HGNetV2 network to replace the backbone of the YOLOv8n model. Secondly, the Bidirectional Feature Pyramid Network (BiFPN) is introduced to enhance the feature fusion layer, further optimizing the model's ability to recognize weed features in complex backgrounds. Finally, a lightweight detection head, LiteDetect, suitable for the BiFPN structure, is designed to streamline the model structure and reduce computational load. Experimental results show that compared to the original YOLOv8n model, YOLO-Weed Nano improves mAP by 1%, while reducing the number of parameters, computation, and weights by 63.8%, 42%, and 60.7%, respectively.

PMID:39747358 | DOI:10.1038/s41598-024-84748-8

Sci Rep. 2025 Jan 2;15(1):179. doi: 10.1038/s41598-024-83090-3.

ABSTRACT

Apprehension of drug action mechanism is paramount for drug response prediction and precision medicine. The unprecedented development of machine learning and deep learning algorithms has expedited the drug response prediction research. However, existing methods mainly focus on forward encoding of drugs, which is to obtain an accurate prediction of the response levels, but omitted to decipher the reaction mechanism between drug molecules and genes. We propose the eXplainable Graph-based Drug response Prediction (XGDP) approach that achieves a precise drug response prediction and reveals the comprehensive mechanism of action between drugs and their targets. XGDP represents drugs with molecular graphs, which naturally preserve the structural information of molecules and a Graph Neural Network module is applied to learn the latent features of molecules. Gene expression data from cancer cell lines are incorporated and processed by a Convolutional Neural Network module. A couple of deep learning attribution algorithms are leveraged to interpret interactions between drug molecular features and genes. We demonstrate that XGDP not only enhances the prediction accuracy compared to pioneering works but is also capable of capturing the salient functional groups of drugs and interactions with significant genes of cancer cells.

PMID:39747341 | DOI:10.1038/s41598-024-83090-3

Sci Rep. 2025 Jan 2;15(1):382. doi: 10.1038/s41598-024-84581-z.

ABSTRACT

Personalized tourism has recently become an increasingly popular mode of travel. Effective personalized route recommendations must consider numerous complex factors, including the vast historical trajectory of tourism, individual traveler preferences, and real-time environmental conditions. However, the large temporal and spatial spans of trajectory data pose significant challenges to achieving high relevance and accuracy in personalized route recommendation systems. This study addresses these challenges by proposing a personalized tourism route recommendation model, the Temporal Multilayer Sequential Neural Network (TMS-Net). The fixed-length trajectory segmentation method designed in TMS-Net can adaptively adjust the segmentation length of tourist trajectories, effectively addressing the issue of large spatiotemporal spans by integrating tourist behavior characteristics and route complexity. The self-attention mechanism incorporating relative positional information enhances the model's ability to capture the relationships between different paths within a tourism route by merging position encoding and distance information. Additionally, the multilayer Long Short-Term Memory neural network module, built through hierarchical time series modeling, deeply captures the complex temporal dependencies in travel routes, improving the relevance of the recommendation results and the ability to recognize long-duration travel behaviors. The TMS-Net model was trained on over six million trajectory data points from Chengdu City, Sichuan Province, spanning January 2016 to December 2022. The experimental results indicated that the optimal trajectory segmentation interval ranged from 0.8 to 1.2 h. The model achieved a recommendation accuracy of 88.6% and a Haversine distance error of 1.23, demonstrating its ability to accurately identify tourist points of interest and provide highly relevant recommendations. This study demonstrates the potential of TMS-Net to improve personalized tourism experiences significantly and offers new methodological insights for personalized travel recommendations.

PMID:39747325 | DOI:10.1038/s41598-024-84581-z

Sci Rep. 2025 Jan 2;15(1):163. doi: 10.1038/s41598-024-84051-6.

ABSTRACT

The last decades have brought an interest in ultrasound applications in dermatology. Especially in the case of atopic dermatitis, where the formation of a subepidermal low echogenic band (SLEB) may serve as an independent indicator of the effects of treatment, the use of ultrasound is of particular interest. This study proposes and evaluates the computer-aided diagnosis method for assessing atopic dermatitis (AD). The fully automated image processing framework combines advanced machine learning techniques for fast, reliable, and repeatable HFUS image analysis, supporting clinical decisions. The proposed methodology comprises accurate SLEB segmentation followed by a classification step. The data set includes 20 MHz images of 80 patients diagnosed with AD according to Hanifin and Rajka criteria, which were evaluated before and after treatment. The ground true labels- clinical evaluation based on Investigator Global Assessment index (IGA score) together with ultrasound skin examination was performed. For reliable analysis, in further experiments, two experts annotated the HFUS images twice in two-week intervals. The analysis aimed to verify whether the fully automated method can classify the HFUS images at the expert level. The Dice coefficient values for segmentation reached 0.908 for SLEB and 0.936 for the entry echo layer. The accuracy of SLEB presence detection results (IGA0) is equal to 98% and slightly outperforms the experts' assessment, which reaches 96%. The overall accuracy of the AD assessment was equal to 69% (Cohen's kappa 0.78) and was comparable with the experts' assessment, ranging between 64% and 70% (Cohen's kappa 0.73-0.79). The results indicate that the automated method can be applied to AD assessment, and its combination with standard diagnosis may benefit repeatable analysis and a better understanding of the processes that take place within the skin and aid treatment monitoring.

PMID:39747292 | DOI:10.1038/s41598-024-84051-6

Sci Rep. 2025 Jan 2;15(1):139. doi: 10.1038/s41598-024-83948-6.

ABSTRACT

Carotid plaques-the buildup of cholesterol, calcium, cellular debris, and fibrous tissues in carotid arteries-can rupture, release microemboli into the cerebral vasculature and cause strokes. The likelihood of a plaque rupturing is thought to be associated with its composition (i.e. lipid, calcium, hemorrhage and inflammatory cell content) and the mechanical properties of the plaque. Automating and digitizing histopathological images of these plaques into tissue specific (lipid and calcified) regions can help us compare histologic findings to in vivo imaging and thereby enable us to optimize medical treatments or interventions for patients based on the composition of plaques. Lack of public datasets and the hypocellular nature of plaques have made applying deep learning to this task difficult. To address this, we sampled 1944 regions of interests from 323 whole slide images and drastically varied their pixel resolution from [Formula: see text] to [Formula: see text] as we anticipated that varying the pixel resolution of histology images can provide neural networks more 'context' that pathologists also rely on. We were able to train Mask R-CNN using regions of interests with varied pixel resolution, with a [Formula: see text] increase in pixel accuracy versus training with patches. The model achieved F1 scores of [Formula: see text] for calcified regions, [Formula: see text] for lipid core with fibrinous material and cholesterol crystals, and [Formula: see text] for fibrous regions, as well as a pixel accuracy of [Formula: see text]. While the F1 score was not calculated for lumen, qualitative results illustrate the model's ability to predict lumen. Hemorrhage was excluded as a class since only one out of 34 carotid endarterectomy specimens had sufficient hemorrhage for annotation.

PMID:39747244 | DOI:10.1038/s41598-024-83948-6