Search Results (32)

This paper proposes a multi-objective collaborative optimization method based on a two-layer optimization framework to address the problem of difficult coordinated optimization of multi-parameter coupling in the frequency regulation performance of heating units under deep peak shaving conditions. The upper-level optimization of this method focuses on the dynamic performance of primary frequency modulation and improves the fast response capability through multi-objective optimization of overshoot and adjustment time. Lower-level optimization is based on the optimal control parameter set output by the upper level, with comprehensive power deviation as the indicator, focusing on suppressing the deviation of frequency modulation power and the steady-state deviation of heating power. Propose a comprehensive quantitative index for frequency modulation performance and characterize the optimization effect of frequency modulation performance. Introducing a dynamic perturbation factor mechanism to generate an improved HO algorithm for dual-layer optimization solutions, preventing it from getting stuck in local optima and solving the problem of global search capability imbalance. The effectiveness of the method was verified based on actual unit calculations, and the obtained control parameter set met the objectives of optimal primary frequency regulation dynamic performance and optimal comprehensive power deviation performance, significantly improving the frequency regulation performance of heating units under deep peak shaving. After optimization, the overshoot performance score of the unit increased by 16.9%, the regulation time performance score increased by 25.1%, the frequency modulation power deviation score increased by 14.2%, the heating power deviation score increased by 17.7%, and the total frequency modulation performance score increased from 75.26 to 95.95, with a comprehensive optimization range of 27.5%. Full article

Black fungus cultivation in bagged form requires frequent inspection of mycelial growth, a process that is labor-intensive and susceptible to subjective judgment. In addition, timely detection of contamination in low-light and high-humidity environments remains a significant challenge. To address these issues, this paper proposed an intelligent agriculture system for black fungus cultivation, with emphasis on practical deployment under real farming conditions. The system integrates a private 5G network with a YOLOv8-based deep learning model for real-time object detection and growth monitoring. Continuous image acquisition and data feedback are achieved through a multi-parameter environmental sensing module and an autonomous ground vehicle (AGV) equipped with IP cameras. To improve model robustness, more than 42,000 labeled training images were generated through data augmentation, and a modified C2f network architecture was employed. Experimental results show that the model achieved a detection accuracy of 93.7% with an average confidence score of 0.96 under live testing conditions. The deployed 5G network provided a downlink throughput of 645.2 Mbps and an uplink throughput of 147.5 Mbps, ensuring sufficient bandwidth and low latency for real-time inference and transmission. Field trials conducted over five cultivation batches demonstrated improvements in disease detection, reductions in labor requirements, and an increase in the average yield success rate to 80%. These findings indicate that the proposed method offers a scalable and practical solution for precision agriculture, integrating next-generation communication technologies with deep learning to enhance cultivation management. Full article

Background/Objectives: This study aimed to develop a reliable scoring system combining clinical and radiological parameters to distinguish atypical lipomatous tumours (ALTs) from lipomas, improving diagnostic accuracy and reducing expensive molecular pathology testing. Methods: A retrospective analysis of 188 patients who underwent surgery for lipomatous tumours was conducted. Patient data, including medical history, pathology, and MRI imaging results, were reviewed. Four predictive models were developed using various clinical and imaging parameters, including age, tumour size, location, and MRI characteristics (homogeneity, contrast enhancement). Statistical analysis, including ROC curve analysis and logistic regression, was performed to assess the accuracy of these models. Results: The highest predictive accuracy was achieved with Model 1, which included seven parameters, yielding an AUC of 0.952. This model achieved a sensitivity of 96.4% and a negative predictive value (NPV) of 97.2%. Reducing the number of parameters lowered the accuracy, with contrast enhancement playing a significant role in Model 1. A risk calculator based on the optimal model was developed, offering an effective tool for clinical use that can be provided. Notably, 21 out of 37 ALTs lacked atypia and would have been missed without molecular testing. Conclusions: The developed scoring system, based on clinical and imaging parameters, accurately distinguishes ALTs from lipomas, offering a practical alternative to molecular pathology testing. This multi-parameter approach significantly improves diagnostic reliability, reducing the risk of misclassification and false negatives, while also potentially lowering healthcare costs. Full article

In the field of active safety technology for intelligent connected vehicles (ICVs), the reliability and safety of the Automatic Emergency Braking (AEB) system is recognized as critical to driving safety. However, existing evaluation methods have been constrained by the inadequacy of static weight assessments in adapting to diverse driving conditions, as well as by the disconnect between conventional evaluation frameworks and experimental validation. To address these limitations, a comprehensive Vehicle-in-the-Loop (VIL) evaluation system based on the dynamic weight analytic hierarchy process (DWAHP) was proposed in this study. A two-tier dynamic weighting architecture was established. At the criterion level, a bivariate variable–weight function, incorporating the vehicle speed and road surface adhesion coefficient, was developed to enable the dynamic coupling modeling of road environment parameters. At the scheme level, a five-dimensional indicator system—integrating braking distance, collision speed, and other key metrics—was constructed to support an adaptive evaluation model under multi-condition scenarios. By establishing a dynamic mapping between weight functions and driving condition parameters, the DWAHP methodology effectively overcame the limitations associated with fixed-weight mechanisms in varying operating conditions. Based on this framework, a dedicated AEB system performance test platform was designed and developed. Validation was conducted using both VIL simulations and real-world road tests, with a Volvo S90L as the test vehicle. The experimental results demonstrated high consistency between VIL and real-world road evaluations across three dimensions: safety (deviation: 0.1833/9.5%), reliability (deviation: 0.2478/13.1%), and riding comfort (deviation: 0.05/2.7%), with an overall comprehensive score deviation of 0.0707 (relative deviation: 0.51%). This study not only verified the technical advantages of the dynamic weight model in adapting to complex driving environments and analyzing multi-parameter coupling effects but also established a systematic methodological framework for evaluating AEB system performance via VIL. The findings provide a robust foundation for the testing and assessment of AEB system, offer a structured approach to advancing the performance evaluation of advanced driver assistance systems (ADASs), facilitate the safe and reliable validation of ICVs’ commercial applications, and ultimately contribute to enhancing road traffic safety. Full article

Facing engineering challenges of real-time and high-precision recognition of strongly heterogeneous formations during directional drilling, it is crucial to address the issues of sparse lithology geological labels and multi-source lithology identification from LWD data. This paper proposes a real-time intelligent recognition method for strongly heterogeneous formations based on multi-parameter logging while drilling and drilling engineering, which can effectively guide directional drilling operations. Traditional supervised learning methods rely heavily on extensive lithology labels and rich wireline logging data. However, in LWD applications, challenges such as limited sample sizes and stringent real-time requirements make it difficult to achieve the accuracy needed for effective geosteering in strongly heterogeneous reservoirs, thereby impacting the reservoir penetration rate. In this study, we comprehensively utilize LWD parameters (six types, including natural gamma and electrical resistivity, etc.) and drilling engineering parameters (four types, including drilling rate and weight on bit, etc.) from offset wells. The UMAP algorithm is employed for nonlinear dimensionality reduction, which not only integrates the dynamic response characteristics of drilling parameters but also preserves the sensitivity of logging data to lithological variations. The K-means clustering algorithm is employed to extract the deep geo-engineering characteristics from multi-source LWD data, thereby constructing a lithology label library and categorizing the training and testing datasets. The optimized CatBoost machine learning model is subsequently utilized for lithology classification, enabling real-time and high-precision geological evaluation during directional drilling. In the Hugin Formation of the Volve field in the Norwegian North Sea, the application of UMAP demonstrates superior data separability compared with PCA and t-SNE, effectively distinguishing thin reservoirs with strong heterogeneity. The CatBoost model achieves a balanced accuracy of 92.7% and an F1-score of 89.3% in six lithology classifications. This approach delivers high-precision geo-engineering decision support for the real-time control of horizontal well trajectories, which holds significant implications for the precision drilling of strongly heterogeneous reservoirs. Full article

Recommender systems aspire to provide users with recommendations that have a high probability of being accepted. This is accomplished by producing rating predictions for products that the users have not evaluated, and, afterwards, the products with the highest prediction scores are recommended to them. Collaborative filtering is a popular recommender system technique which generates rating prediction scores by blending the ratings that users with similar preferences have previously given to these products. However, predictions may entail errors, which will either lead to recommending products that the users would not accept or failing to recommend products that the users would actually accept. The first case is considered much more critical, since the recommender system will lose a significant amount of reliability and consequently interest. In this paper, after performing a study on rating prediction confidence factors in collaborative filtering, (a) we introduce the concept of prediction reliability classes, (b) we rank these classes in relation to the utility of the rating predictions belonging to each class, and (c) we present a collaborative filtering recommendation algorithm which exploits these reliability classes for prediction formulation. The efficacy of the presented algorithm is evaluated through an extensive multi-parameter evaluation process, which demonstrates that it significantly enhances recommendation quality. Full article

Objectives: Prediction of lung function deficits following pulmonary infection is challenging and suffers from inaccuracy. We sought to develop machine-learning models for prediction of post-inflammatory lung changes based on COVID-19 recovery data. Methods: In the prospective CovILD study (n = 420 longitudinal observations from n = 140 COVID-19 survivors), data on lung function testing (LFT), chest CT including severity scoring by a human radiologist and density measurement by artificial intelligence, demography, and persistent symptoms were collected. This information was used to develop models of numeric readouts and abnormalities of LFT with four machine learning algorithms (Random Forest, gradient boosted machines, neural network, and support vector machines). Results: Reduced DLCO (diffusion capacity for carbon monoxide <80% of reference) was found in 94 (22%) observations. Those observations were modeled with a cross-validated accuracy of 82–85%, AUC of 0.87–0.9, and Cohen’s κ of 0.45–0.5. No reliable models could be established for FEV1 or FVC. For DLCO as a continuous variable, three machine learning algorithms yielded meaningful models with cross-validated mean absolute errors of 11.6–12.5% and R² of 0.26–0.34. CT-derived features such as opacity, high opacity, and CT severity score were among the most influential predictors of DLCO impairment. Conclusions: Multi-parameter machine learning trained with demographic, clinical, and artificial intelligence chest CT data reliably and reproducibly predicts LFT deficits and outperforms single markers of lung pathology and human radiologist’s assessment. It may improve diagnostic and foster personalized treatment. Full article

Background: Glioblastoma is an aggressive and incurable type of brain cancer. Little progress has been made in the development of effective new therapies in the past decades. The blood–brain barrier (BBB) and drug efflux pumps, which together hamper drug delivery to these tumors, play a pivotal role in the gap between promising preclinical findings and failure in clinical trials. Therefore, selecting drugs that can reach the tumor region in pharmacologically effective concentrations is of major importance. Methods: In the current study, we utilized a drug selection platform to identify candidate drugs by combining in vitro oncological drug screening data and pharmacokinetic (PK) profiles for central nervous system (CNS) penetration using the multiparameter optimization (MPO) score. Furthermore, we developed intracranial patient-derived xenograft (PDX) models that recapitulated the in situ characteristics of glioblastoma and characterized them in terms of vascular integrity, BBB permeability and expression of ATP-binding cassette (ABC) transporters. Omacetaxine mepesuccinate (OMA) was selected as a proof-of-concept drug candidate to validate our drug selection pipeline. Results: We assessed OMA’s PK profile in three different orthotopic mouse PDX models and found that OMA reaches the brain tumor tissue at concentrations ranging from 2- to 11-fold higher than in vitro IC₅₀ values on patient-derived glioblastoma cell cultures. Conclusions: This study demonstrates that OMA, a drug selected for its in vitro anti-glioma activity and CNS- MPO score, achieves brain tumor tissue concentrations exceeding its in vitro IC₅₀ values in patient-derived glioblastoma cell cultures, as shown in three orthotopic mouse PDX models. We emphasize the importance of such approaches at the preclinical level, highlighting both their significance and limitations in identifying compounds with potential clinical implementation in glioblastoma. Full article

Magnetic resonance imaging (MRI) has been proven to be an indispensable imaging method in bladder cancer, and it can accurately identify muscular invasion of bladder cancer. Multiparameter MRI is a promising tool widely used for preoperative staging evaluation of bladder cancer. Vesical Imaging-Reporting and Data System (VI-RADS) scoring has proven to be a reliable tool for local staging of bladder cancer with high accuracy in preoperative staging, but VI-RADS still faces challenges and needs further improvement. Artificial intelligence (AI) holds great promise in improving the accuracy of diagnosis and predicting the prognosis of bladder cancer. Automated machine learning techniques based on radiomics features derived from MRI have been utilized in bladder cancer diagnosis and have demonstrated promising potential for practical implementation. Future work should focus on conducting more prospective, multicenter studies to validate the additional value of quantitative studies and optimize prediction models by combining other biomarkers, such as urine and serum biomarkers. This review assesses the value of multiparameter MRI in the accurate evaluation of muscular invasion of bladder cancer, as well as the current status and progress of its application in the evaluation of efficacy and prognosis. Full article

The association between plasma amyloid-beta protein (Aβ) and subjective cognitive decline (SCD) remains controversial. We aimed to explore the correlation between neuroimaging findings, plasma Aβ, and neuropsychological scales using data from 53 SCD patients and 46 age- and sex-matched healthy controls (HCs). Magnetic resonance imaging (MRI) was used to obtain neuroimaging data for a whole-brain voxel-based morphometry analysis and cortical functional network topological features. The SCD group had slightly lower Montreal Cognitive Assessment (MoCA) scores than the HC group. The Aβ42 levels were significantly higher in the SCD group than in the HC group (p < 0.05). The SCD patients demonstrated reduced volumes in the left hippocampus, right rectal gyrus (REC.R), and right precentral gyrus (PreCG.R); an increased percentage fluctuation in the left thalamus (PerAF); and lower average small-world coefficient (aSigma) and average global efficiency (aEg) values. Correlation analyses with Aβ and neuropsychological scales revealed significant positive correlations between the volumes of the HIP.L, REC.R, PreCG.R, and MoCA scores. The HIP.L volume and Aβ42 were negatively correlated, as were the REC.R volume and Aβ42/40. PerAF and aSigma were negatively and positively correlated with the MoCA scores, respectively. The aEg was positively correlated with Aβ42/40. SCD patients may exhibit alterations in plasma biomarkers and multi-parameter MRI that resemble those observed in Alzheimer’s disease, offering a theoretical foundation for early clinical intervention in SCD. Full article

Prostate-specific acid phosphatase (PSAP) is a marker for prostate cancer. To assess the specificity and prognostic impact of PSAP, 14,137 samples from 127 different tumor (sub)types, 17,747 prostate cancers, and 76 different normal tissue types were analyzed via immunohistochemistry in a tissue microarray format. In normal tissues, PSAP staining was limited to the prostate epithelial cells. In prostate cancers, PSAP was seen in 100% of Gleason 3 + 3, 95.5% of Gleason 4 + 4, 93.8% of recurrent cancer under androgen deprivation therapy, 91.0% of Gleason 5 + 5, and 31.2% of small cell neuroendocrine cancer. In non-prostatic tumors, PSAP immunostaining was only found in 3.2% of pancreatic neuroendocrine tumors and in 0.8% of diffuse-type gastric adenocarcinomas. In prostate cancer, reduced PSAP staining was strongly linked to an advanced pT stage, a high classical and quantitative Gleason score, lymph node metastasis, high pre-operative PSA levels, early PSA recurrence (p < 0.0001 each), high androgen receptor expression, and TMPRSS2:ERG fusions. A low level of PSAP expression was linked to PSA recurrence independent of pre- and postoperative prognostic markers in ERG-negative cancers. Positive PSAP immunostaining is highly specific for prostate cancer. Reduced PSAP expression is associated with aggressive prostate cancers. These findings make PSAP a candidate marker for prognostic multiparameter panels in ERG-negative prostate cancers. Full article

Purpose: The aim of this study was to ascertain whether radiomics data can assist in differentiating small (<4 cm) clear cell renal cell carcinomas (ccRCCs) from small oncocytomas using T2-weighted magnetic resonance imaging (MRI). Material and Methods: This retrospective study incorporated 48 tumors, 28 of which were ccRCCs and 20 were oncocytomas. All tumors were less than 4 cm in size and had undergone pre-biopsy or pre-surgery MRI. Following image pre-processing, 102 radiomics features were evaluated. A univariate analysis was performed using the Wilcoxon rank-sum test with Bonferroni correction. We compared multiple radiomics pipelines of normalization, feature selection, and machine learning (ML) algorithms, including random forest (RF), logistic regression (LR), AdaBoost, K-nearest neighbor, and support vector machine, using a supervised ML approach. Results: No statistically significant features were identified via the univariate analysis with Bonferroni correction. The most effective algorithm was identified using a pipeline incorporating standard normalization, RF-based feature selection, and LR, which achieved an area under the curve (AUC) of 83%, accuracy of 73%, sensitivity of 79%, and specificity of 65%. Subsequently, the most significant features were identified from this algorithm, and two groups of uncorrelated features were established based on Pearson correlation scores. Using these features, an algorithm was established after a pipeline of standard normalization and LR, achieving an AUC of 90%, an accuracy of 77%, sensitivity of 83%, and specificity of 69% for distinguishing ccRCCs from oncocytomas. Conclusions: Radiomics analysis based on T2-weighted MRI can aid in distinguishing small ccRCCs from small oncocytomas. However, it is not superior to standard multiparameter renal MRI and does not yet allow us to dispense with percutaneous biopsy. Full article

Background: The International Prognostic Index (IPI) is applied to predict the outcome of chronic lymphocytic leukemia (CLL) with five prognostic factors, including genetic analysis. We investigated whether multiparameter flow cytometry (MPFC) data of CLL samples could predict the outcome by methods of explainable artificial intelligence (XAI). Further, XAI should explain the results based on distinctive cell populations in MPFC dot plots. Methods: We analyzed MPFC data from the peripheral blood of 157 patients with CLL. The ALPODS XAI algorithm was used to identify cell populations that were predictive of inferior outcomes (death, failure of first-line treatment). The diagnostic ability of each XAI population was evaluated with receiver operating characteristic (ROC) curves. Results: ALPODS defined 17 populations with higher ability than the CLL-IPI to classify clinical outcomes (ROC: area under curve (AUC) 0.95 vs. 0.78). The best single classifier was an XAI population consisting of CD4+ T cells (AUC 0.78; 95% CI 0.70–0.86; p < 0.0001). Patients with low CD4+ T cells had an inferior outcome. The addition of the CD4+ T-cell population enhanced the predictive ability of the CLL-IPI (AUC 0.83; 95% CI 0.77–0.90; p < 0.0001). Conclusions: The ALPODS XAI algorithm detected highly predictive cell populations in CLL that may be able to refine conventional prognostic scores such as IPI. Full article

Background: Intratumor microbiomes can influence tumorigenesis and progression. The relationship between intratumor microbiomes and cervical cancer metastasis, however, remains unclear. Methods: We examined 294 cervical cancer samples together with information on microbial expression, identified metastasis-associated microbiomes, and used machine learning methods to validate their predictive ability on tumor metastasis. The tumors were subsequently typed based on differences in microbial expression. Differentially expressed genes in different tumor types were combined to construct a tumor-prognostic risk score model and a multiparameter nomogram model. In addition, we performed a functional enrichment analysis of differentially expressed genes to infer the mechanism of action between microbiomes and tumor cells. Results: Based on the 15 differentially expressed microbiomes, machine learning models were able to correctly predict the risk of cervical cancer metastasis. In addition, both the risk score and the nomogram model accurately predicted tumor prognosis. Differences in the expression of endogenous genes in tumors can influence the distribution of the intracellular microbiomes. Conclusions: Intratumoral microbiomes in cervical cancer are associated with tumor metastasis and influence disease prognosis. A change in gene expression within tumor cells is responsible for differences in the microbial populations within the tumor. Full article

Cytotoxic T-cells (CTLs) exhibit strong effector functions to leverage antigen-specific anti-tumoral and anti-viral immunity. When naïve CTLs are activated by antigen-presenting cells (APCs) they display various levels of functional heterogeneity. To investigate this, we developed a single-cell droplet microfluidics platform that allows for deciphering single CTL activation profiles by multi-parameter analysis. We identified and correlated functional heterogeneity based on secretion profiles of IFNγ, TNFα, IL-2, and CD69 and CD25 surface marker expression levels. Furthermore, we strengthened our approach by incorporating low-melting agarose to encapsulate pairs of single CTLs and artificial APCs in hydrogel droplets, thereby preserving spatial information over cell pairs. This approach provides a robust tool for high-throughput and single-cell analysis of CTLs compatible with flow cytometry for subsequent analysis and sorting. The ability to score CTL quality, combined with various potential downstream analyses, could pave the way for the selection of potent CTLs for cell-based therapeutic strategies. Full article