Biomed Tech (Berl). 2025 Jan 29. doi: 10.1515/bmt-2024-0282. Online ahead of print.

ABSTRACT

OBJECTIVES: In recent years, significant progress has been made in the research of gesture recognition using surface electromyography (sEMG) signals based on machine learning and deep learning techniques. The main motivation for sEMG gesture recognition research is to provide more natural, convenient, and personalized human-computer interaction, which makes research in this field have considerable application prospects in rehabilitation technology. However, the existing gesture recognition algorithms still need to be further improved in terms of global feature capture, model computational complexity, and generalizability.

METHODS: This paper proposes a fusion model of Squeeze-and-Excitation Networks (SE) and DenseNet, inserting attention mechanism between DenseBlock and Transition to focus on the most important information, improving feature representation ability, and effectively solving the problem of gradient vanishing.

RESULTS: This proposed method was tested on the electromyographic gesture datasets NinaPro DB2 and DB4, achieving accuracies of 85.93 and 82.39 % respectively. Through ablation experiments, it was found that the method based on DenseNet-101 as the backbone model produced the best results.

CONCLUSIONS: Compared with existing models, this proposed method has better robustness and generalizability in gesture recognition, providing new ideas for the development of sEMG signal gesture recognition applications in the future.

PMID:39873377 | DOI:10.1515/bmt-2024-0282

IET Syst Biol. 2025 Jan-Dec;19(1):e70001. doi: 10.1049/syb2.70001.

ABSTRACT

Metal ions are significant ligands that bind to proteins and play crucial roles in cell metabolism, material transport, and signal transduction. Predicting the protein-metal ion ligand binding residues (PMILBRs) accurately is a challenging task in theoretical calculations. In this study, the authors employed fused amino acids and their derived information as feature parameters to predict PMILBRs using three classical machine learning algorithms, yielding favourable prediction results. Subsequently, deep learning algorithm was incorporated in the prediction, resulting in improved results for the sets of Ca2+ and Mg2+ compared to previous studies. The validation matrix provided the optimal prediction model for each ionic ligand binding residue, exhibiting the capability of effectively predicting the binding sites of metal ion ligands for real protein chains.

PMID:39873344 | DOI:10.1049/syb2.70001

J Biophotonics. 2025 Jan 28:e202400557. doi: 10.1002/jbio.202400557. Online ahead of print.

ABSTRACT

Liver malignancies, particularly hepatocellular carcinoma (HCC), pose a formidable global health challenge. Conventional diagnostic techniques frequently fall short in precision, especially at advanced HCC stages. In response, we have developed a novel diagnostic strategy that integrates hyperspectral imaging with deep learning. This innovative approach captures detailed spectral data from tissue samples, pinpointing subtle cellular differences that elude traditional methods. A sophisticated deep convolutional neural network processes this data, effectively distinguishing high-grade liver cancer from cirrhosis with an accuracy of 89.45%, a sensitivity of 90.29%, and a specificity of 88.64%. For HCC differentiation specifically, it achieves an impressive accuracy of 93.73%, sensitivity of 92.53%, and specificity of 90.07%. Our results underscore the potential of this technique as a precise, rapid, and non-invasive diagnostic tool that surpasses existing clinical methods in staging liver cancer and differentiating cirrhosis.

PMID:39873135 | DOI:10.1002/jbio.202400557

Front Immunol. 2025 Jan 13;15:1479458. doi: 10.3389/fimmu.2024.1479458. eCollection 2024.

ABSTRACT

INTRODUCTION: Diffuse parenchymal lung diseases (DPLD) cover heterogeneous types of lung disorders. Among many pathological phenotypes, pulmonary fibrosis is the most devastating and represents a characteristic sign of idiopathic pulmonary fibrosis (IPF). Despite a poor prognosis brought by pulmonary fibrosis, there are no specific diagnostic biomarkers for the initial development of this fatal condition. The major hallmark of lung fibrosis is uncontrolled activation of lung fibroblasts to myofibroblasts associated with extracellular matrix deposition and the loss of both lung structure and function.

METHODS: Here, we used this peculiar feature in order to identify specific biomarkers of pulmonary fibrosis in bronchoalveolar lavage fluids (BALF). The primary MRC-5 human fibroblasts were activated with BALF collected from patients with clinically diagnosed lung fibrosis; the activated fibroblasts were then washed rigorously, and further incubated to allow secretion. Afterwards, the secretomes were analysed by mass spectrometry.

RESULTS: In this way, the CD44 protein was identified; consequently, BALF of all DPLD patients were positively tested for the presence of CD44 by ELISA. Finally, biochemical and biophysical characterizations revealed an exosomal origin of CD44. Receiver operating characteristics curve analysis confirmed CD44 in BALF as a specific and reliable biomarker of IPF and other types of DPLD accompanied with pulmonary fibrosis.

PMID:39872532 | PMC:PMC11769834 | DOI:10.3389/fimmu.2024.1479458

J Med Chem. 2025 Jan 28. doi: 10.1021/acs.jmedchem.4c02415. Online ahead of print.

ABSTRACT

Molecular glue degraders (MGDs) are small molecules that facilitate proximity between a target protein and an E3 ubiquitin ligase, thereby inducing target protein degradation. Glutarimide-containing compounds are MGDs that bind cereblon (CRBN) and recruit neosubstrates. Through explorative synthesis of a glutarimide-based library, we discovered a series of molecules that induce casein kinase 1 alpha (CK1α) degradation. By scaffold hopping and rational modification of the chemical scaffold, we identified an imidazo[1,2-a]pyrimidine compound that induces potent and selective CK1α degradation. A structure-activity relationship study of the lead compound, QXG-6442, identified the chemical features that contribute to degradation potency and selectivity compared to other frequently observed neosubstrates. The glutarimide library screening and structure-activity relationship medicinal chemistry approach we employed is generally useful for developing new molecular glue degraders toward new targets of interest.

PMID:39873536 | DOI:10.1021/acs.jmedchem.4c02415

Bio Protoc. 2025 Jan 20;15(2):e5163. doi: 10.21769/BioProtoc.5163. eCollection 2025 Jan 20.

ABSTRACT

Nicotinic acetylcholine receptors (nAChRs) are a family of ligand-gated ion channels expressed in nervous and non-nervous system tissue important for memory, movement, and sensory processes. The pharmacological targeting of nAChRs, using small molecules or peptides, is a promising approach for the development of compounds for the treatment of various human diseases including inflammatory and neurogenerative disorders such as Alzheimer's disease. Using the Aplysia californica acetylcholine binding protein (Ac-AChBP) as an established structural surrogate for human homopentameric α7 nAChRs, we describe an innovative protein painting mass spectrometry (MS) method that can be used to identify interaction sites for various ligands at the extracellular nAChR site. We describe how the use of small molecule dyes can be optimized to uncover contact sites for ligand-protein interactions based on MS detection. Protein painting MS has been recently shown to be an effective tool for the identification of residues within Ac-AChBP involved in the binding of know ligands such as α-bungarotoxin. This strategy can be used with computational structural modeling to identify binding regions involved in drug targeting at the nAChR. Key features • Identify binding ligands of nicotinic receptors based on similarity with the acetylcholine binding protein. • Can be adapted to test various ligands and binding conditions. • Mass spectrometry identification of specific amino acid residues that contribute to protein binding. • Can be effectively coupled to structural modeling analysis.

PMID:39872720 | PMC:PMC11769746 | DOI:10.21769/BioProtoc.5163

Bio Protoc. 2025 Jan 20;15(2):e5169. doi: 10.21769/BioProtoc.5169. eCollection 2025 Jan 20.

ABSTRACT

Primary neuronal culture and transient transfection offer a pair of crucial tools for neuroscience research, providing a controlled environment to study the behavior, function, and interactions of neurons in vitro. These cultures can be used to investigate fundamental aspects of neuronal development and plasticity, as well as disease mechanisms. There are numerous methods of transient transfection, such as electroporation, calcium phosphate precipitation, or cationic lipid transfection. In this protocol, we used electroporation for neurons immediately before plating and cationic lipid transfection for neurons that have been cultured for a few days in vitro. In our experience, the transfection efficiency of electroporation can be as high as 30%, and cationic lipid transfection has an efficiency of 1%-2%. While cationic lipid transfection has much lower efficiency than electroporation, it does offer the advantage of a higher expression level. Therefore, these transfection methods are suitable for different stages of neurons and different expression requirements. Key features • Culture of primary neurons from the CNS. • Electroporation for freshly isolated neurons in suspension. • Cationic lipid transfection for adherent neurons.

PMID:39872712 | PMC:PMC11769751 | DOI:10.21769/BioProtoc.5169

Life Med. 2024 Nov 25;3(6):lnae039. doi: 10.1093/lifemedi/lnae039. eCollection 2024 Dec.

NO ABSTRACT

PMID:39872152 | PMC:PMC11761737 | DOI:10.1093/lifemedi/lnae039

Ophthalmol Ther. 2025 Jan 28. doi: 10.1007/s40123-024-01089-5. Online ahead of print.

ABSTRACT

INTRODUCTION: Lotilaner ophthalmic solution (0.25%) is the first United States Food and Drug Administration (US FDA)-approved drug for treating Demodex blepharitis. In pivotal trials, it was found to be well tolerated and demonstrated a significant reduction in collarettes and mite density after a 6-week treatment regimen. This study aimed to report the safety and efficacy profile of lotilaner ophthalmic solution (0.25%) from a pooled analysis of two pivotal trials in patients with Demodex blepharitis.

METHODS: Pooled data were analyzed from two randomized, double-masked, vehicle-controlled clinical trials [phase 2b/3 Saturn-1 (NCT04475432) and phase 3 Saturn-2 (NCT04784091)] in which patients with Demodex blepharitis were randomly assigned in a 1:1 ratio to receive either lotilaner ophthalmic solution (0.25%) (study group) or the vehicle formulation without lotilaner (control group), twice daily for 6 weeks. The outcome measures were the proportion of patients with 0-2 collarettes (grade 0 collarettes), mite eradication, erythema cure, and the proportion of patients with ≤ 10 collarettes (grade 0 or 1 collarettes) at day 43.

RESULTS: Overall, 833 participants were randomized to receive either the study drug (N = 415) or vehicle (N = 418). On day 43, 49.8% of patients in the study group vs. 9.9% in the control group (p < 0.0001) had collarette grade 0 (0-2 collarettes). A reduction to ≤ 10 collarettes (grade 0 or 1 collarettes) was achieved in 85.1% of patients in study group vs. 28.0% in control group (p < 0.0001). The proportion of patients achieving mite eradication (60.2% vs. 16.1%, p < 0.0001) and erythema cure (24.9% vs. 7.9%, p < 0.0001) were also statistically significantly higher in the study group compared to the control group. The rates of adverse events were low in both studies, with no serious drug-related ocular adverse events reported. As many as 92% of patients rated the study drop as neutral to very comfortable.

CONCLUSIONS: Twice-daily treatment with lotilaner ophthalmic solution (0.25%) for 6 weeks demonstrated statistical significance for all outcome measures compared to the vehicle control, with low rates of adverse events and a high rate of drop comfort.

PMID:39873946 | DOI:10.1007/s40123-024-01089-5

Clin Kidney J. 2024 Dec 3;18(1):sfae394. doi: 10.1093/ckj/sfae394. eCollection 2025 Jan.

ABSTRACT

BACKGROUND: Sparsentan, a dual-acting antagonist for both the angiotensin II receptor type 1 and the endothelin receptor type A, has emerged as a promising therapeutic agent for the treatment of IgA nephropathy (IgAN). Following the publication of the PROTECT trial, sparsentan recently received approval for the treatment of IgAN in Europe. However, it remains uncertain whether an additive effect can be observed in the context of existing treatment with sodium-glucose co-transporter 2 (SGLT2) inhibitors, given that the PROTECT study did not investigate this dual therapy approach.

METHODS: A total of 23 patients with IgAN were treated with sparsentan via the Managed Access Programme between December 2023 and August 2024. The patients were stable on maximum tolerated doses of renin-angiotensin system (RAS) and SGLT2 inhibitors, with an estimated glomerular filtration rate (eGFR) >30 mL/min/1.73 m² and a urine protein/creatinine ratio (UPCR) >0.75 g/g.

RESULTS: In the 23 patients, median (IQR) baseline eGFR (CKD-EPI) was 42 mL/min/1.73 m2 (32-63) and median baseline UPCR was 1.5 g/g (0.9-1.8). After initiation of sparsentan, UPCR significantly decreased (P < 0.0001) to a median of 0.85 g/g (0.42-1.15) in the 2-week follow-up and further declined (P = 0.001) to a median of 0.60 g/g (0.32-0.82) after 14 weeks, equivalent to a relative reduction in proteinuria up to 62% (45-74). A similar significant reduction was observed for the urine albumin/creatinine ratio. No drug-related serious adverse events were reported.

CONCLUSIONS: In this real-world setting, sparsentan shows a significant impact on proteinuria, leading to a relative reduction of 62% in UPCR after 14 weeks and beyond, even in patients already receiving SGLT2 inhibitors.

PMID:39872637 | PMC:PMC11770278 | DOI:10.1093/ckj/sfae394

bioRxiv [Preprint]. 2025 Jan 17:2025.01.15.633240. doi: 10.1101/2025.01.15.633240.

ABSTRACT

Deep learning sequence models trained on personalized genomics can improve variant effect prediction, however, applications of these models are limited by computational requirements for storing and reading large datasets. We address this with GenVarLoader, which stores personalized genomic data in new memory-mapped formats with optimal data locality to achieve ~1,000x faster throughput and ~2,000x better compression compared to existing alternatives.

PMID:39868273 | PMC:PMC11761601 | DOI:10.1101/2025.01.15.633240

bioRxiv [Preprint]. 2025 Jan 13:2025.01.13.632696. doi: 10.1101/2025.01.13.632696.

ABSTRACT

Autophagy is an essential cellular process which functions to maintain homeostasis in response to stressors such as starvation or infection. Here, we report that a subset of autophagy factors including ATG-3 play an antiviral role in Orsay virus infection of Caenorhabditis elegans. Orsay virus infection does not modulate autophagic flux, and re-feeding after starvation limits Orsay virus infection and blocks autophagic flux, suggesting that the role of ATG-3 in Orsay virus susceptibility is independent of its role in maintaining autophagic flux. atg-3 mutants phenocopy rde-1 mutants, which have a defect in RNA interference (RNAi), in susceptibility to Orsay virus infection and transcriptional response to infection. However, atg-3 mutants do not exhibit defects in RNAi. Additionally, atg-3 limits viral infection at a post-entry step, similar to rde-1 mutants. Differential expression analysis using RNA sequencing revealed that antiviral sqt-2, which encodes a collagen trimer protein, is depleted in naïve and infected atg-3 mutants, as well as in infected WT animals, as are numerous other collagen genes. These data suggest that ATG-3 has a role in collagen organization pathways that function in antiviral defense in C. elegans.

PMID:39868230 | PMC:PMC11761658 | DOI:10.1101/2025.01.13.632696

bioRxiv [Preprint]. 2025 Jan 14:2025.01.13.632860. doi: 10.1101/2025.01.13.632860.

ABSTRACT

Success of phage therapies is limited by bacterial defenses against phages. While a large variety of anti-phage defense mechanisms has been characterized, how expression of these systems is distributed across individual cells and how their combined activities translate into protection from phages has not been studied. Using bacterial single-cell RNA sequencing, we profiled the transcriptomes of ~50,000 cells from cultures of a human pathobiont, Bacteroides fragilis, infected with a lytic bacteriophage. We quantified the asynchronous progression of phage infection in single bacterial cells and reconstructed the infection timeline, characterizing both host and phage transcriptomic changes as infection unfolded. We discovered a subpopulation of bacteria that remained uninfected and determined the heterogeneously expressed host factors associated with protection. Each cell's vulnerability to phage infection was defined by combinatorial phase-variable expression of multiple genetic loci, including capsular polysaccharide (CPS) biosynthesis pathways, restriction-modification systems (RM), and a previously uncharacterized operon likely encoding fimbrial genes. By acting together, these heterogeneously expressed phase-variable systems and anti-phage defense mechanisms create a phenotypic landscape where distinct protective combinations enable the survival and re-growth of bacteria expressing these phenotypes without acquiring additional mutations. The emerging model of complementary action of multiple protective mechanisms heterogeneously expressed across an isogenic bacterial population showcases the potent role of phase variation and stochasticity in bacterial anti-phage defenses.

PMID:39868116 | PMC:PMC11761130 | DOI:10.1101/2025.01.13.632860

Sci Rep. 2025 Jan 27;15(1):3344. doi: 10.1038/s41598-025-88050-z.

ABSTRACT

In cybersecurity, anomaly detection in tabular data is essential for ensuring information security. While traditional machine learning and deep learning methods have shown some success, they continue to face significant challenges in terms of generalization. To address these limitations, this paper presents an innovative method for tabular data anomaly detection based on large language models, called "Tabular Anomaly Detection via Guided Prompts" (TAD-GP). This approach utilizes a 7-billion-parameter open-source model and incorporates strategies such as data sample introduction, anomaly type recognition, chain-of-thought reasoning, multi-turn dialogue, and key information reinforcement. Experimental results indicate that the TAD-GP framework improves F1 scores by 79.31%, 97.96%, and 59.09% on the CICIDS2017, KDD Cup 1999, and UNSW-NB15 datasets, respectively. Furthermore, the smaller-scale TAD-GP model outperforms larger models across multiple datasets, demonstrating its practical potential in environments with constrained computational resources and requirements for private deployment. This method addresses a critical gap in research on anomaly detection in cybersecurity, specifically using small-scale open-source models.

PMID:39870811 | DOI:10.1038/s41598-025-88050-z

Sci Rep. 2025 Jan 27;15(1):3391. doi: 10.1038/s41598-025-86073-0.

ABSTRACT

The Sharp-van der Heijde score (SvH) is crucial for assessing joint damage in rheumatoid arthritis (RA) through radiographic images. However, manual scoring is time-consuming and subject to variability. This study proposes a multistage deep learning model to predict the Overall Sharp Score (OSS) from hand X-ray images. The framework involves four stages: image preprocessing, hand segmentation with UNet, joint identification via YOLOv7, and OSS prediction utilizing a custom Vision Transformer (ViT). Evaluation metrics included Intersection over Union (IoU), Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), Huber loss, and Intraclass Correlation Coefficient (ICC). The model was trained using stratified group 3-fold cross-validation on a dataset of 679 patients and tested externally on 291 subjects. The joint identification model achieved 99% accuracy. The ViT model achieved the best OSS prediction for patients with Sharp scores < 50. It achieved a Huber loss of 4.9, an RMSE of 9.73, and an MAE of 5.35, demonstrating a strong correlation with expert scores (ICC = 0.702, P < 0.001). This study is the first to apply a ViT for OSS prediction in RA. It presents an efficient and automated alternative for overall damage assessment. This approach may reduce reliance on manual scoring.

PMID:39870749 | DOI:10.1038/s41598-025-86073-0

Sci Rep. 2025 Jan 27;15(1):3389. doi: 10.1038/s41598-025-87370-4.

ABSTRACT

With the emergence of numerous classifications, surgical treatment for adolescent idiopathic scoliosis (AIS) can be guided more effectively. However, surgical decision-making and optimal strategies still lack standardization and personalized customization. Our study aims to devise proper deep learning (DL) models that incorporate key factors influencing surgical outcomes on the coronal plane in AIS patients to facilitate surgical decision-making and predict surgical results for AIS patients. A total of 425 AIS patients who underwent posterior spinal fixation were collected. Variables such as age, gender, preoperative and final follow-up horizontal and vertical coordinate vectors, and screw positioning data were preprocessed by parameterizing image data and transforming various data types into a unified, continuous high-dimensional feature space. Four deep learning models were designed, including Multi-Layer Perceptron model, Encoder-Decoder model, CNN-LSTM Attention model and Deep FM model. For the implementation of deep learning, 70% of the data was adopted for training and 30% for evaluation. The mean square error (MSE), mean absolute error (MAE) and curve fitting between the predicted and corresponding real postoperative spinal coordinates of the test set were adopted to validate and compare the efficacy of the DL models. A total of 425 patients with an average age of 14.60 ± 2.08 years, including 77 males and 348 females, were enrolled in this study. The Lenke type 1 and 5 AIS patients accounted for the majority of the included patients. The results showed that the Multi-Layer Perceptron model achieved the best performance among the four DL models, with a mean square error of 2.77 × 10-5 and an average absolute error of 0.00350 on the validation set. Moreover, the results predicted by the Multi-Layer Perceptron model closely matched the actual coordinate positions on the original postoperative images of patients with Lenke type 1 and AIS patients. Deep learning models can provide alternative and effective decision-making support for AIS patients undergoing surgery. Regarding the learning curve and data volume, the optimal DL models should be adjusted and refined to meet future demands.

PMID:39870730 | DOI:10.1038/s41598-025-87370-4

Sci Rep. 2025 Jan 27;15(1):3383. doi: 10.1038/s41598-024-84749-7.

ABSTRACT

To retrospectively develop and validate an interpretable deep learning model and nomogram utilizing endoscopic ultrasound (EUS) images to predict pancreatic neuroendocrine tumors (PNETs). Following confirmation via pathological examination, a retrospective analysis was performed on a cohort of 266 patients, comprising 115 individuals diagnosed with PNETs and 151 with pancreatic cancer. These patients were randomly assigned to the training or test group in a 7:3 ratio. The least absolute shrinkage and selection operator algorithm was employed to reduce the dimensionality of deep learning (DL) features extracted from pre-standardized EUS images. The retained nonzero coefficient features were subsequently applied to develop predictive eight DL models based on distinct machine learning algorithms. The optimal DL model was identified and used to establish a clinical signature, which subsequently informed the construction and evaluation of a nomogram. Gradient-weighted Class Activation Mapping (Grad-CAM) and Shapley Additive Explanations (SHAP) were implemented to interpret and visualize the model outputs. A total of 2048 DL features were initially extracted, from which only 27 features with coefficients greater than zero were retained. The support vector machine (SVM) DL model demonstrated exceptional performance, achieving area under the curve (AUC) values of 0.948 and 0.795 in the training and test groups, respectively. Additionally, a nomogram was developed, incorporating both DL and clinical signatures, and was visually represented for practical application. Finally, the calibration curves, decision curve analysis (DCA) plots, and clinical impact curves (CIC) exhibited by the DL model and nomogram indicated high accuracy. The application of Grad-CAM and SHAP enhanced the interpretability of these models. These methodologies contributed substantial net benefits to clinical decision-making processes. A novel interpretable DL model and nomogram were developed and validated using EUS images, cooperating with machine learning algorithms. This approach demonstrates significant potential for enhancing the clinical applicability of EUS in predicting PNETs from pancreatic cancer, thereby offering valuable insights for future research and implementation.

PMID:39870667 | DOI:10.1038/s41598-024-84749-7

Clin Transl Sci. 2025 Feb;18(2):e70059. doi: 10.1111/cts.70059.

ABSTRACT

Tramadol, the 41st most prescribed drug in the United States in 2021 is a prodrug activated by CYP2D6, which is highly polymorphic. Previous studies showed enzyme-inhibitor affinity varied between different CYP2D6 allelic variants with dextromethorphan and atomoxetine metabolism. However, no study has compared tramadol metabolism in different CYP2D6 alleles with different CYP2D6 inhibitors. We hypothesize that the inhibitory effects of CYP2D6 inhibitors on CYP2D6-mediated tramadol metabolism are inhibitor- and CYP2D6-allele-specific. We performed comparative analyses of CYP2D6*1, CYP2D6*2, CYP2D6*10, and CYP2D6*17 using recombinant enzymes to metabolize tramadol to O-desmethyltramadol, measured via UPLC-MS/MS. The Michaelis constant (Km) and maximum velocity (Vmax) for each CYP2D6 allele, and IC50 values for different inhibitors were determined by nonlinear regression analysis. Intrinsic clearance was calculated as Vmax/Km. The intrinsic clearance of tramadol was almost double for CYP2D6*2 (180%) but was much lower for CYP2D6*10 and *17 (20% and 10%, respectively) compared to CYP2D6*1. The inhibitor potencies (defined by Ki) for the various inhibitors for the CYP2D6*1 allele were quinidine > terbinafine > paroxetine ≈ duloxetine >>bupropion. CYP2D6*2 showed the next greatest inhibition, with Ki ratios compared to CYP2D6*1 ranging from 0.96 to 3.87. For each inhibitor tested, CYP2D6*10 and CYP2D6*17 were more resistant to inhibition than CYP2D6*1 or CYP2D6*2, with most Ki ratios in the 3-9 range. Three common CYP2D6 allelic variants showed different metabolic capacities toward tramadol and genotype-dependent inhibition compared to CYP2D6*1. Further studies are warranted to understand the clinical consequences of inhibitor and CYP2D6 genotype-dependent drug-drug interactions on tramadol bioactivation.

PMID:39870079 | DOI:10.1111/cts.70059

Med Mycol. 2025 Jan 27:myaf006. doi: 10.1093/mmy/myaf006. Online ahead of print.

ABSTRACT

This study was performed to evaluate whether the MIC Test Strip (MTS) quantitative assay for determining the minimum inhibitory concentration (MIC) correlated with the CLSI reference broth microdilution method (BMD) for antifungal susceptibility testing of wild-type and non-wild-type Aspergillus species isolated from cystic fibrosis patients against antifungal agents known to be usually effective against Aspergillus spp. This study was performed to assist in the decision-making process for possible deployment of the MTS assay for antimicrobial susceptibility testing of Aspergillus species into regional public health laboratories of Mycology due to difficulties in equipping the reference BMD methods in a laboratory routine. For this purpose, a set of 40 phenotypically diverse isolates (27 wild-type, 9 non-wild-type, and 4 species with reduced susceptibility to azoles and amphotericin B (AMB)) collected from clinical samples were tested. MICs were performed by both MTS and reference BMD for AMB, and azoles. MTS results for posaconazole correlated well with reference BMD rendering an almost perfect agreement (kappa value = 1.000) by category interpretation (CI)/category distribution of MICs (CDM) (100%) while voriconazole MTS results yielded a substantial correlation with BMD (kappa value = 0.788) by CI/CDM (97.5%). In contrast, itraconazole and AMB yielded the poorest correlation with BMD, rendering a moderate agreement (kappa values of 0.554 and 0.437, respectively) by CI/CDM (87.5% and 85%, respectively). In conclusion, the MTS method represents a valid option for antimicrobial susceptibility testing of Aspergillus species against posaconazole and voriconazole. Itraconazole and AMB MTS results showed some concerning lack of correlation with the corresponding reference BMD results.

PMID:39870380 | DOI:10.1093/mmy/myaf006

Med Phys. 2025 Jan 27. doi: 10.1002/mp.17639. Online ahead of print.

ABSTRACT

BACKGROUND: Kidney tumors, common in the urinary system, have widely varying survival rates post-surgery. Current prognostic methods rely on invasive biopsies, highlighting the need for non-invasive, accurate prediction models to assist in clinical decision-making.

PURPOSE: This study aimed to construct a K-means clustering algorithm enhanced by Transformer-based feature transformation to predict the overall survival rate of patients after kidney tumor resection and provide an interpretability analysis of the model to assist in clinical decision-making.

METHODS: This study was based on a publicly available C4KC-KiTS-2019 dataset from the TCIA database, including preoperative computed tomography (CT) images and survival time data of 210 patients. Initially, the radiomics features of the kidney tumor area were extracted using the 3D slicer software. Feature selection was then conducted using ICC, mRMR algorithms, and LASSO regression to calculate radiomics scores. Subsequently, the selected features were input into a pre-trained Transformer model for feature transformation to obtain a higher-dimensional feature set. Then, K-means clustering was performed using this feature set, and the model was evaluated using receiver operating characteristic (ROC) and Kaplan-Meier curves. Finally, the SHAP interpretability algorithm was used for the feature importance analysis of the K-means clustering results.

RESULTS: Eleven important features were selected from 851 radiomics features. The K-means clustering model after Transformer feature transformation showed AUCs of 0.889, 0.841, and 0.926 for predicting 1-, 3-, and 5-year overall survival rates, respectively, thereby outperforming both the K-means model with original feature inputs and the radiomics score method. A clustering analysis revealed survival prognosis differences among different patient groups, and a SHAP analysis provided insights into the features that had the most significant impacts on the model predictions.

CONCLUSIONS: The K-means clustering algorithm enhanced by the Transformer feature transformation proposed in this study demonstrates promising accuracy and interpretability in predicting the overall survival rate after kidney tumor resection. This method provides a valuable tool for clinical decision-making and contributes to improved management and treatment strategies for patients with kidney tumors.

PMID:39871101 | DOI:10.1002/mp.17639