Expert Rev Proteomics. 2025 Feb 16. doi: 10.1080/14789450.2025.2468300. Online ahead of print.

ABSTRACT

INTRODUCTION: Rare diseases (RDs) are a heterogeneous group of diseases recognized as a relevant global health priority but posing aspects of complexity such as: geographical scattering of affected individuals, improper/late diagnosis, limited awareness, difficult surveillance and monitoring, limited understanding of natural history, and lack of treatment. Usually, RDs have a pediatric onset and are life-long, multisystemic, and associated with a poor prognosis.

AREAS COVERED: In this work, we review how high-throughput omics technologies such as genomics, transcriptomics, proteomics, metabolomics, epigenomics, and other well-established omics, which are increasingly more affordable and efficient, can be applied to the study of RDs promoting diagnosis, understanding of pathological mechanisms, biomarker discovery and identification of treatments.

EXPERT OPINION: RDs, despite their challenges, offer a niche where collaborative efforts and personalized treatment strategies might be feasible using omics technologies. Specialized consortia fostering multidisciplinary collaboration, data sharing, and the development of biobanks and registries can be built; multi-omics approaches, including so far less exploited omics technologies, along with the implementation of AI tools can be undertaken to deepen our understanding of RDs, driving biomarker discovery and clinical interventions. Nevertheless, technical, ethical, legal and societal issues must be clearly defined and addressed.

PMID:39956998 | DOI:10.1080/14789450.2025.2468300

Am J Transl Res. 2025 Jan 15;17(1):685-692. doi: 10.62347/EHTT9544. eCollection 2025.

ABSTRACT

OBJECTIVE: This study aimed to assess the efficacy and safety of sacral nerve root magnetic stimulation (SNRMS) combined with solifenacin in female patients with overactive bladder (OAB) symptoms.

METHODS: A total of 183 female patients with OAB symptoms were prospectively randomized into 2 groups. Ninety-two patients in the combination group accepted SNRMS and solifenacin therapy and 91 patients serving as control accepted only solifenacin therapy. The lower urinary tract symptoms, OAB questionnaire (OAB-q) health-related quality of life (HRQoL), symptom bother score, and overactive bladder syndrome score (OABSS) were compared between the two groups at the end of the second, fourth, and eighth weeks.

RESULTS: The incidence of lower urinary tract symptoms, including urgency, frequent urination, and incontinence episodes, was significantly lower in the fourth and eighth weeks in patients of the combination treatment group than those in the solifenacin group (P < 0.05). The incidence of drug-related adverse events in the two groups was similar, with no statistically significant difference (P > 0.05). The OAB-q HRQoL score in the combination group was significantly higher than that in the solifenacin group between the fourth and eighth weeks (P < 0.05). Meanwhile, the OAB-q symptom bother score and OABSS were both lower in the combination group than those in the solifenacin group from the fourth to eighth weeks (P < 0.05).

CONCLUSIONS: The combination therapy of SNRMS and solifenacin demonstrated significant improvements over solifenacin monotherapy in reducing OAB symptoms in female patients, providing a higher QoL without increasing bothersome adverse effects.

PMID:39959226 | PMC:PMC11826160 | DOI:10.62347/EHTT9544

Liver Res. 2024 Oct 24;8(4):295-303. doi: 10.1016/j.livres.2024.10.001. eCollection 2024 Dec.

ABSTRACT

BACKGROUND AND AIMS: Antiviral therapy is essential for hepatitis B virus-related acute-on-chronic liver failure (HBV-ACLF). No data are available on the long-term prognosis or safety of tenofovir alafenamide (TAF), tenofovir disoproxil fumarate (TDF), or entecavir (ETV) in treating HBV-ACLF globally. This study was conducted to investigate the long-term efficacy and safety of the three nucleos(t)ide analogs in the treatment of HBV-ACLF.

METHODS: In this prospective, real-world cohort study, patients with HBV-ACLF were assigned to the TAF, TDF, and ETV groups. A total of 199 patients completed the 144-week follow-up. After propensity score matching (PSM), 44 patients remained in each group for further analysis of survival status, incidence of hepatocellular carcinoma (HCC), virological response, and liver and renal function indicators.

RESULTS: In the original cohort, HCC developed in one patient in each group. No serious drug-related adverse events were observed. In the PSM cohort, the 144-week survival rates were 56.82%, 75.00%, and 59.09% in the TAF, TDF, and ETV groups, respectively (P = 0.118). When stratified into noncirrhosis and cirrhosis subgroups at baseline, the survival rate of the ETV group was slightly lower than that of the TAF and TDF group in noncirrhosis patients (P = 0.338), and the survival rate of the TAF group was slightly lower than that of the TDF and ETV group in cirrhosis patients (P = 0.052), but the differences were not statistically significant. The long-term overall survival rates in the TAF, TDF, and ETV groups were comparable. After 144 weeks, no significant difference in the virological response rate or liver or renal function indicators was found among the three groups, except for the level of aspartate aminotransferase, which was significantly higher in the TDF group than in the ETV group at week 144 (P = 0.001).

CONCLUSIONS: There were no significant differences in the survival rate, incidence of HCC, efficacy or safety associated with the use of these three nucleos(t)ide analogs in treating HBV-ACLF.

TRIAL REGISTRATION: ClinicalTrials.gov NCT03920618.

PMID:39958923 | PMC:PMC11771274 | DOI:10.1016/j.livres.2024.10.001

Discoveries (Craiova). 2024 Dec 31;12(4):e201. doi: 10.15190/d.2024.20. eCollection 2024 Oct-Dec.

ABSTRACT

Lomotil (diphenoxylate-atropine) toxicity in the pediatric population remains a significant concern particularly in low and lower middle-income countries. This may result from accidental ingestion or inappropriate therapeutic administration which can lead to life threatening complications including respiratory and central nervous system depression. A 2-year-old child presented to the pediatric emergency room in an altered state of consciousness. Clinical examination revealed dry mucous membranes, and a prolonged capillary refill time with weak radial pulses. Keeping in view the one-day history of 10-12 episodes of acute onset loose, watery stools, patient was initially treated as a case of hypovolemic shock. With rehydration therapy, his perfusion improved. However, the Glasgow Coma Scale score remained 8, as was observed on initial presentation. Upon further probing, it was revealed by the parents that the child had been given Lomotil by a local general practitioner for unresolved watery diarrhea. Pinpoint pupils and slow shallow vesicular breathing confirmed this diagnosis of Lomotil overdose. Administration of 0.1mg/kg/dose Naloxone repeated once, completely reversed the toxic effects. The child was able to make a full recovery and was discharged the following day. This case highlights the importance of recognizing and managing diphenoxylate toxicity in children, emphasizing the need for increased clinical awareness. A lack of consensus regarding the toxic dose of this drug reveals a gap warranting further research and establishment of standardized guidelines to ensure accurate dosing and improved patient safety.

PMID:39958914 | PMC:PMC11830492 | DOI:10.15190/d.2024.20

Phys Med. 2025 Feb 15;131:104935. doi: 10.1016/j.ejmp.2025.104935. Online ahead of print.

ABSTRACT

PURPOSE: To assess the lower [18F]FDG limit in administered activity and/or scan time reduction capabilities of a digital-BGO 32-cm axial field-of-view PET system while being compliant with current and updated EANM Research Ltd Fluorine-18 accreditation specifications (EARL1 and EARL2).

METHODS: EARL1 and EARL2 compliance of the digital-BGO system (Omni Legend 32 cm) was tested for several reconstructions, including those that apply precision deep learning-based image enhancement (PDL) as postprocessing, using the calibration QC and NEMA IEC phantom measurements. The image quality QC scan was repeated every hour for 7 h, with each subsequent hour representing a lower administered activity, and reconstructed for various times per bed position, i.e. 30, 60, 120, 180, and 300 s. For each of the image quality QC images, coefficient of variation (COV) of the background compartment, and mean, maximum and peak activity concentration recovery coefficients (RCmean, RCmax and RCpeak) of differently-sized spheres were calculated and compared to current and updated EARL accreditation specifications.

RESULTS: When we apply 1 min per bed position for PET acquisition, [18F]FDG administration can be reduced by a factor of ∼ 4 for EARL1, by a factor of ∼ 8 for EARL2 (2 mm voxels) and by a factor of ∼ 4 for EARL2 (4 mm voxels) using both standard reconstructions and PDL post-processing compared to current EANM recommendations for [18F]FDG administration (7 MBqminbed-1kg-1).

CONCLUSIONS: Reduction in [18F]FDG administered activity is possible by at least a factor 4 for 1 min/bed with the Omni Legend 32 cm PET/CT while maintaining EARL1 and EARL2 compliance.

PMID:39956005 | DOI:10.1016/j.ejmp.2025.104935

BMJ Open. 2025 Feb 16;15(2):e089418. doi: 10.1136/bmjopen-2024-089418.

ABSTRACT

INTRODUCTION: Rare disease treatments (RDTs) promise considerable patient benefit but the evidence to demonstrate their value in health technology assessment (HTA) is often limited. HTA outcomes for RDTs vary across countries and there are differences in how uncertainty is dealt with by HTA agencies. Yet, there is limited comparative research assessing how different HTA agencies consider issues affecting evidence quality and uncertainty in RDT appraisals. This protocol describes a systematic and consistent approach for data extraction from RDT appraisal documents produced to inform decisions by HTA agencies. By documenting data extraction rules transparently, we ensure reproducibility and reliability of analyses of the extracted data.

METHODS AND ANALYSIS: We will select RDT appraisals issued by the National Institute for Health and Care Excellence (NICE) in England and the Federal Joint Committee (GBA) in Germany, using predefined inclusion criteria. We will extract data from appraisal documents in accordance with the rules set out in this protocol. We will analyse the extracted data to investigate how issues affecting evidence quality and uncertainty as documented in appraisals are considered, highlighting the similarities and differences between countries and identifying factors that are associated with HTA outcomes.

ETHICS AND DISSEMINATION: This study was approved by the Ethics Committee of the London School of Hygiene & Tropical Medicine (reference number 29156). Study results will be submitted for publication in peer-reviewed journals.

PMID:39956595 | DOI:10.1136/bmjopen-2024-089418

Ultrason Sonochem. 2025 Feb 12;114:107271. doi: 10.1016/j.ultsonch.2025.107271. Online ahead of print.

ABSTRACT

Javanese turmeric (Curcuma xanthorrhiza Roxb.) is known for its diverse pharmacological activities due to its rich phytoconstituents, including curcuminoids and xanthorrhizol. Typically, these compounds are extracted using organic solvents, which pose health and environmental risks. Therefore, safer and more environmentally friendly green extraction methods are being developed. This study investigated the effect of ultrasound-assisted extraction (UAE) combined with natural deep eutectic solvents (NADES) based on choline chloride and organic acids (lactic, malic, and citric acid) to find the best combination for extracting curcuminoids and xanthorrhizol from Javanese turmeric. Results showed that UAE using choline chloride and malic acid (1:1) (ChCl-MA) yielded the best results. The Box-Behnken Design optimized water addition, solvent-to-powder ratio, and extraction time, with optimal conditions being 25 % water addition, a 20 mL/g ratio, and a 15-minute extraction time. This method yielded 4.58 mg/g of curcuminoids and 12.93 mg/g of xanthorrhizol. Furthermore, the ChCl-MA NADES with UAE extraction showed more cytoselective activity towards the HeLa cancer cell line compared to the non-cancer HaCaT cell line. In contrast, traditional ethanol extraction was non-selective, as indicated by similar cell viability reductions in both HeLa and HaCaT cells at 6.25 ppm. Collectively, this study is the first to report the optimal NADES combination with UAE, based on salts and organic acids, for the extraction of Javanese turmeric rhizomes with selective cytotoxic effects against cancer cells. These findings may contribute to the development of novel, naturally derived anticancer agents using green extraction techniques.

PMID:39955874 | DOI:10.1016/j.ultsonch.2025.107271

J Cyst Fibros. 2025 Feb 16:S1569-1993(25)00054-2. doi: 10.1016/j.jcf.2025.02.005. Online ahead of print.

ABSTRACT

BACKGROUND: Despite the existence of numerous clinical practice guidelines (CPGs) for cystic fibrosis (CF), there is limited understanding of their credibility and consistency. This systematic review aims to comprehensively evaluate the quality of CPGs for CF and its pulmonary complications, focusing on treatment recommendations for pulmonary care.

METHODS: We conducted a comprehensive search across four databases and relevant websites to identify eligible guidelines providing treatment recommendations. The quality of these guidelines was assessed using the Appraisal of Guidelines for Research and Evaluation (AGREE) II tool. Pulmonary treatment recommendations were analyzed and synthesized narratively.

RESULTS: A total of 35 guidelines were identified. Most guidelines were of moderate quality according to the AGREE II instrument, with overall scores ranging from 21·05 to 76·13. Only six guidelines were recommended for use. These guidelines provide 359 pulmonary treatment recommendations for seven primary therapies and others. There was inconsistency in the use of airway clearance therapy, anti-inflammatories, antibiotics, inhaled drugs, and cystic fibrosis transmembrane conductance regulator modulator therapy. Four guidelines conditionally advocated for oral corticosteroids, while six opposed routine inhaled corticosteroids. One guideline discouraged lumacaftor-ivacaftor in the general CF population, two recommended only for children under 12 years old, and another strongly advocated for children between 2 and 5 years of age. However, one guideline noted a lack of evidence to recommend it for children under 6.

CONCLUSION: The quality of CPGs for CF and its pulmonary complications has improved over time, reaching a moderate level generally, but there is still room for further improvement. Future efforts should focus on standardizing methodological frameworks and generating robust clinical evidence to enhance the overall quality and applicability of CF guidelines.

PMID:39956717 | DOI:10.1016/j.jcf.2025.02.005

J Cyst Fibros. 2025 Feb 15:S1569-1993(25)00055-4. doi: 10.1016/j.jcf.2025.02.003. Online ahead of print.

ABSTRACT

BACKGROUND: In individuals with cystic fibrosis (CF), respiratory viral infections frequently result in hospitalization and have been linked to secondary bacterial infection and colonization, highlighting viral infections as possible contributors to CF lung disease progression. We hypothesized that expression of antiviral host defense genes is dysregulated in CF airway epithelia.

METHODS: We infected primary CF and Non-CF airway epithelia with respiratory syncytial virus (RSV) and characterized their responses at 12 hr, 24 hr, 48 hr, 72 hr, and 120 hr post infection (hpi) by RNA sequencing (RNAseq).

RESULTS: Our analysis revealed strikingly different gene expression profiles for the CF and Non-CF epithelia over the course of the infection. While both CF and Non-CF cells exhibited an early signature for interferon signaling and antiviral defense pathways, this response was relatively exaggerated and sustained in CF epithelia. We also observed, in both genotypes, a transient downregulation of cilia-associated genes and loss of ciliary activity by 72 hpi. Interestingly, recovery of cilia activity was delayed in the CF epithelia.

CONCLUSIONS: These findings further our understanding of innate immune dysfunction in the CF airway epithelium and suggest that virus-induced cilia injury may further compromise host defenses in CF airways.

PMID:39956716 | DOI:10.1016/j.jcf.2025.02.003

J Cyst Fibros. 2025 Feb 15:S1569-1993(25)00052-9. doi: 10.1016/j.jcf.2025.02.001. Online ahead of print.

ABSTRACT

BACKGROUND: Cystic fibrosis (CF) has been associated with impaired cardiovascular and endothelial function. CF transmembrane conductance regulator (CFTR) modulator therapy, most recently Elexacaftor/Tezacaftor/Ivacaftor (ETI), has led to improved CFTR function and life expectancy. However, the rising prevalence of obesity in adults is concerning. This study assessed the micro- and macrovascular endothelial function, cardiovascular disease (CVD) risk factors, and physical activity (PA) profiles in people with CF (pwCF) on ETI compared to healthy matched controls.

METHODS: In 15 pwCF and 15 age- and sex-matched controls, microvascular endothelial function (via transdermal delivery of insulin [INS] and acetylcholine [ACh] on the forearm), macrovascular endothelial function (via flow-mediated dilation [FMD] of the brachial artery), central haemodynamic parameters, including heart rate (HR), stroke volume index (SVi) and cardiac output index (Q̇I) (via thoracic impedance cardiography), body mass index (BMI), blood pressure (BP), and accelerometer-assessed PA were measured.

RESULTS: There were no differences in INS or FMD-mediated vasodilation between the groups (P > 0.05). However, a reduced vasodilatory response was evident in pwCF following ACh-mediated vasodilation (P = 0.01) and FMD normalised for shear rate (P = 0.03). No differences in resting HR, SVi, Q̇I, BP, BMI or PA were found (P > 0.05).

CONCLUSION: This study demonstrated reduced micro- and macrovascular function in pwCF. This dysfunction may have potential health implications, particularly regarding long-term cardiovascular risk and further longitudinal assessments are warranted.

PMID:39956715 | DOI:10.1016/j.jcf.2025.02.001

J Cyst Fibros. 2025 Feb 15:S1569-1993(25)00060-8. doi: 10.1016/j.jcf.2025.02.007. Online ahead of print.

ABSTRACT

BACKGROUND & AIMS: Cystic fibrosis hepatobiliary involvement is a heterogeneous and systemic entity. The primary objective was to determine the prevalence of steatosis, by magnetic resonance-proton density fat fraction (MR-PDFF), and liver fibrosis, by magnetic resonance elastography (MRE), in a cohort of adults with cystic fibrosis. The secondary objective was to determine the diagnostic yield of widely available non-invasive liver markers for steatosis and fibrosis, and vibration controlled transitional elastography (VCTE) releasing Control Attenuation Parameter (CAP) (dB/m) and stiffness (kPa), with the aim of proposing a diagnostic algorithm.

METHODS: We conducted a cross-sectional study including 101 adult patients with cystic fibrosis seen in a multidisciplinary unit. The study encompassed a clinical evaluation, morpho-functional assessment, VCTE, non-invasive liver markers and MR-PDFF and MRE. Diagnostic accuracy was assessed using ROC curves and 2 × 2 tables.

RESULTS: MR-PDFF detected hepatic steatosis in 18 of 101 (17.8 %) patients, while MRE detected significant liver fibrosis in 15 of 101 (14.9 %). The VCTE cut-off with the best diagnostic yield, determined by the Youden index, was 222 dB/m for the presence of steatosis (AUC 0.864 (95 % CI 0.768-0.961; p < 0.001) and the VCTE cut-off was 6.65 kPa for liver fibrosis (AUC 0.951(95 % CI 0.81-1; p = 0.053). A screening algorithm for hepatic steatosis was developed using the fatty liver index (FLI) and CAP, with a negative predictive value of 83.3 %. For liver fibrosis, it was outperformed by the Hepamet Fibrosis Score (HFS) and VCTE, with a negative predictive value of 100 %.

CONCLUSIONS: The prevalence of hepatic steatosis and liver fibrosis was 17.8 % and 14.9 %, respectively. VCTE alone or in combination with FLI for steatosis or HFS for fibrosis demonstrated high diagnostic accuracy. This approach effectively allows for the exclusion of steatosis and fibrosis, thereby reducing the need for MR-PDFF and MRE studies.

PMID:39956714 | DOI:10.1016/j.jcf.2025.02.007

Cancer Imaging. 2025 Feb 16;25(1):14. doi: 10.1186/s40644-025-00837-5.

ABSTRACT

BACKGROUND: Immunotherapy has revolutionized the treatment landscape for head and neck squamous cell carcinoma (HNSCC) and PD-L1 combined positivity score (CPS) scoring is recommended as a biomarker for immunotherapy. Therefore, this study aimed to develop an MRI-based deep learning score (DLS) to non-invasively assess PD-L1 expression status in HNSCC patients and evaluate its potential effeciency in predicting prognostic stratification following treatment with immune checkpoint inhibitors (ICI).

METHODS: In this study, we collected data from four patient cohorts comprising a total of 610 HNSCC patients from two separate institutions. We developed deep learning models based on the ResNet-101 convolutional neural network to analyze three MRI sequences (T1WI, T2WI, and contrast-enhanced T1WI). Tumor regions were manually segmented, and features extracted from different MRI sequences were fused using a transformer-based model incorporating attention mechanisms. The model's performance in predicting PD-L1 expression was evaluated using the area under the curve (AUC), sensitivity, specificity, and calibration metrics. Survival analyses were conducted using Kaplan-Meier survival curves and log-rank tests to evaluate the prognostic significance of the DLS.

RESULTS: The DLS demonstrated high predictive accuracy for PD-L1 expression, achieving an AUC of 0.981, 0.860 and 0.803 in the training, internal and external validation cohort. Patients with higher DLS scores demonstrated significantly improved progression-free survival (PFS) in both the internal validation cohort (hazard ratio: 0.491; 95% CI, 0.270-0.892; P = 0.005) and the external validation cohort (hazard ratio: 0.617; 95% CI, 0.391-0.973; P = 0.040). In the ICI-treated cohort, the DLS achieved an AUC of 0.739 for predicting durable clinical benefit (DCB).

CONCLUSIONS: The proposed DLS offered a non-invasive and accurate approach for assessing PD-L1 expression in patients with HNSCC and effectively stratified HNSCC patients to benefit from immunotherapy based on PFS.

PMID:39956910 | DOI:10.1186/s40644-025-00837-5

Neurol Res Pract. 2025 Feb 17;7(1):11. doi: 10.1186/s42466-025-00367-2.

ABSTRACT

BACKGROUND: Artificial Intelligence is influencing medicine on all levels. Neurology, one of the most complex and progressive medical disciplines, is no exception. No longer limited to neuroimaging, where data-driven approaches were initiated, machine and deep learning methodologies are taking neurologic diagnostics, prognostication, predictions, decision making and even therapy to very promising potentials.

MAIN BODY: In this review, the basic principles of different types of Artificial Intelligence and the options to apply them to neurology are summarized. Examples of noteworthy studies on such applications are presented from the fields of acute and intensive care neurology, stroke, epilepsy, and movement disorders. Finally, these potentials are matched with risks and challenges jeopardizing ethics, safety and equality, that need to be heeded by neurologists welcoming Artificial Intelligence to their field of expertise.

CONCLUSION: Artificial intelligence is and will be changing neurology. Studies need to be taken to the prospective level and algorithms undergo federated learning to reach generalizability. Neurologists need to master not only the benefits but also the risks in safety, ethics and equity of such data-driven form of medicine.

PMID:39956906 | DOI:10.1186/s42466-025-00367-2

J Transl Med. 2025 Feb 16;23(1):188. doi: 10.1186/s12967-025-06183-1.

ABSTRACT

BACKGROUND: Predicting long-term outcomes in liver transplantation remain a challenging endeavor. This research aims to harness the power of deep learning to develop an advanced prognostic model for assessing long-term outcomes, with a specific focus on distinguishing between deceased and living donor transplantation.

METHODS: A comprehensive dataset from UNOS encompassing clinical, demographic, and transplant-related variables of liver transplant recipients from deceased and living donors was utilized. The main dataset has been transformed into Deceased Donor-Recipient and Living Donor-Recipient dataset. After manual extraction, the dimensionality reduction was performed with Principal component analysis in both datasets and top ranked 23 attributes were collected. A Deeplearning4j Multilayer Perceptron classifier has been employed and long-term survival analysis has been conducted with the help of liver follow-up data. The performance evaluation is done separately in datasets and evaluated the survival probabilities of 23 years.

RESULTS: UNOS database comprises 410 attributes and 353,589 records from 1998 to 2023. The outcome from the deep learning model was compared with actual graft survival to ensure the accuracy. The model trained 23 attributes and obtained Sensitivity, Specificity and accuracy values were 99.9, 99.9 and 99.91% using R-Living donor dataset. The Sensitivity, Specificity and Accuracy value obtained using R-Deceased donor dataset were 99.7, 99.7 and 99.86%. The short term and long-term survival prediction after liver transplantation has been done successfully with Dl4jMLP classifier with appropriate selection of attributes irrespective of donor type. This study's finding suggesting that the distinction between deceased and living donor transplantation does not significantly affect survival prediction after liver transplantation is noteworthy.

CONCLUSIONS: The utility of the Deeplearning4j model in survival prediction after liver transplantation has been validated in this study. Based on the findings, deceased donor transplantation could be promoted over living donor transplantation.

PMID:39956905 | DOI:10.1186/s12967-025-06183-1

Commun Eng. 2025 Feb 16;4(1):22. doi: 10.1038/s44172-025-00370-0.

ABSTRACT

Sports helmets provide incomplete protection against brain injuries. Here we aim to improve helmet liner efficiency by employing a novel approach that optimizes their properties. By exploiting a finite element model that simulates head impacts, we developed deep learning models that predict the peak rotational velocity and acceleration of a dummy head protected by various liner materials. The deep learning models exhibited a remarkable correlation coefficient of 0.99 within the testing dataset with mean absolute error of 0.8 rad.s-1 and 0.6 krad.s-2 respectively, highlighting their predictive ability. Deep learning-based material optimization demonstrated a significant reduction in the risk of brain injuries, ranging from -5% to -65%, for impact energies between 250 and 500 Joules. This result emphasizes the effectiveness of material design to mitigate sport-related brain injury risks. This research introduces promising avenues for optimizing helmet designs to enhance their protective capabilities.

PMID:39956866 | DOI:10.1038/s44172-025-00370-0

Sci Rep. 2025 Feb 16;15(1):5685. doi: 10.1038/s41598-025-89557-1.

ABSTRACT

Glaucoma is characterised by progressive vision loss due to retinal ganglion cell deterioration, leading to gradual visual field (VF) impairment. The standard VF test may be impractical in some cases, where optical coherence tomography (OCT) can offer predictive insights into VF for multimodal diagnoses. However, predicting VF measures from OCT data remains challenging. To address this, five regression models were developed to predict VF measures from OCT, Shapley Additive exPlanations (SHAP) analysis was performed for interpretability, and a clinical software tool called OCT to VF Predictor was developed. To evaluate the models, a total of 268 glaucomatous eyes (86 early, 72 moderate, 110 advanced) and 226 normal eyes were included. The machine learning models outperformed recent OCT-based VF prediction deep learning studies, with correlation coefficients of 0.76, 0.80 and 0.76 for mean deviation, visual field index and pattern standard deviation, respectively. Introducing the pointwise normalisation and step-size concept, a mean absolute error of 2.51 dB was obtained in pointwise sensitivity prediction, and the grayscale prediction model yielded a mean structural similarity index of 77%. The SHAP-based analysis provided critical insights into the most relevant features for glaucoma diagnosis, showing promise in assisting eye care practitioners through an explainable AI tool.

PMID:39956834 | DOI:10.1038/s41598-025-89557-1

Commun Biol. 2025 Feb 16;8(1):247. doi: 10.1038/s42003-025-07694-9.

ABSTRACT

Accurate computational determination of RNA-protein interactions remains challenging, particularly when encountering unknown RNAs and proteins. The limited number of RNAs and their flexibility constrained the effectiveness of the deep-learning models for RNA-protein interaction prediction. Here, we introduce ZHMolGraph, which integrates graph neural network and unsupervised large language models to predict RNA-protein interaction. We validate ZHMolGraph predictions on two benchmark datasets and outperform the current best methods. For the dataset of entirely unknown RNAs and proteins, ZHMolGraph shows an improvement in achieving high AUROC of 79.8% and AUPRC of 82.0%. This represents a substantial improvement of 7.1%-28.7% in AUROC and 4.6%-30.0% in AUPRC over other methods. We utilize ZHMolGraph to enhance the challenging SARS-CoV-2 RPI and unbound RNA-protein complex predictions. Such enhancements make ZHMolGraph a reliable option for genome-wide RNA-protein prediction. ZHMolGraph holds broad potential for modeling and designing RNA-protein complexes.

PMID:39956833 | DOI:10.1038/s42003-025-07694-9

Sci Data. 2025 Feb 16;12(1):276. doi: 10.1038/s41597-025-04454-6.

ABSTRACT

Automated Optical Inspection (AOI) technology is crucial for industrial defect detection but struggles with shadows and surface reflectivity, resulting in false positives and missed detections, especially on non-planar parts. To address these issues, a novel defect detection technique based on deep learning and photometric stereo vision was proposed, along with the creation of the Metal Surface Defect Dataset (MSDD). The proposed Stroboscopic Illuminant Image Acquisition (SIIA) method uses a specially arranged illuminant setup and a Taylor Series Channel Mixer (TSCM) to blend multi-angle illumination images into pseudo-color images. This approach enables end-to-end defect detection using universal object detectors. The method involves mapping color space transformations to spatial domain transformations and utilizing hue randomization for data augmentation. Four object detection methods (FCOS, YOLOv5, YOLOv8, and RT-DETR) were validated on the MSDD, achieving an mAP of 86.1%, surpassing traditional methods. The MSDD includes 138,585 single-channel images and 9,239 mixed images, covering eight defect types. This dataset is essential for automated visual inspection of metal surfaces and is freely accessible for research purposes.

PMID:39956811 | DOI:10.1038/s41597-025-04454-6

Nat Commun. 2025 Feb 16;16(1):1691. doi: 10.1038/s41467-025-57027-x.

ABSTRACT

Single-cell ATAC-seq technology advances our understanding of single-cell heterogeneity in gene regulation by enabling exploration of epigenetic landscapes and regulatory elements. However, low sequencing depth per cell leads to data sparsity and high dimensionality, limiting the characterization of gene regulatory elements. Here, we develop scAGDE, a single-cell chromatin accessibility model-based deep graph representation learning method that simultaneously learns representation and clustering through explicit modeling of data generation. Our evaluations demonstrated that scAGDE outperforms existing methods in cell segregation, key marker identification, and visualization across diverse datasets while mitigating dropout events and unveiling hidden chromatin-accessible regions. We find that scAGDE preferentially identifies enhancer-like regions and elucidates complex regulatory landscapes, pinpointing putative enhancers regulating the constitutive expression of CTLA4 and the transcriptional dynamics of CD8A in immune cells. When applied to human brain tissue, scAGDE successfully annotated cis-regulatory element-specified cell types and revealed functional diversity and regulatory mechanisms of glutamatergic neurons.

PMID:39956806 | DOI:10.1038/s41467-025-57027-x

Acad Radiol. 2025 Feb 15:S1076-6332(25)00101-1. doi: 10.1016/j.acra.2025.01.046. Online ahead of print.

ABSTRACT

RATIONALE AND OBJECTIVES: Immunotherapy combined with chemotherapy has improved outcomes for some esophageal squamous cell carcinoma (ESCC) patients, but accurate pre-treatment risk stratification remains a critical gap. This study constructed a deep learning (DL) model to predict survival outcomes in ESCC patients receiving immunotherapy combined with chemotherapy.

MATERIALS AND METHODS: A DL model was developed to predict survival outcomes in ESCC patients receiving immunotherapy and chemotherapy. Retrospective data from 482 patients across three institutions were split into training (N=322), internal test (N=79), and external test (N=81) sets. Unenhanced computed tomography (CT) scans were processed to analyze tumor and peritumoral regions. The model evaluated multiple input configurations: original tumor regions of interest (ROIs), ROI subregions, and ROIs expanded by 1 and 3 pixels. Performance was assessed using Harrell's C-index and receiver operating characteristic (ROC) curves. A multimodal model combined DL-derived risk scores with five key clinical and laboratory features. The Shapley Additive Explanations (SHAP) method elucidated the contribution of individual features to model predictions.

RESULTS: The DL model with 1-pixel peritumoral expansion achieved the best accuracy, yielding a C-index of 0.75 for the internal test set and 0.60 for the external test set. Hazard ratios for high-risk patients were 1.82 (95% CI: 1.19-2.46; P=0.02) in internal test set. The multimodal model achieved C-indices of 0.74 and 0.61 for internal and external test sets, respectively. Kaplan-Meier analysis revealed significant survival differences between high- and low-risk groups (P<0.05). SHAP analysis identified tumor response, risk score, and age as critical contributors to predictions.

CONCLUSION: This DL model demonstrates efficacy in stratifying ESCC patients by survival risk, particularly when integrating peritumoral imaging and clinical features. The model could serve as a valuable pre-treatment tool to facilitate the implementation of personalized treatment strategies for ESCC patients undergoing immunotherapy and chemotherapy.

PMID:39956748 | DOI:10.1016/j.acra.2025.01.046