ACS Sens. 2025 Feb 27. doi: 10.1021/acssensors.4c03366. Online ahead of print.

ABSTRACT

Flexible wearable potentiometric ion sensors for continuous monitoring of electrolyte cations have made significant advances in bioanalysis for personal healthcare and diagnostics. However, less attention is paid to the most abundant extracellular anion, chloride ion (Cl-) as a mark of electrolyte imbalance and an important diagnostic indicator of cystic fibrosis, which has important significance for accurate monitoring in complex biological fluids. An all-solid-state Cl--selective electrode is constructed utilizing oxygen vacancies reinforced vanadium oxide with a nitrogen-doped carbon shield as the solid contact (V2O3-x@NC/Cl--ISE). The prepared V2O3-x@NC/Cl--ISE exhibits a low detection limit of 10-5.45 M without an interfacial water layer and shows a highly stable potential with 7.24 μV/h during 24 h, which is attributed to the rapid interfacial electron transfer of the conductive carbon layers and the valence state transition of the polyvalent vanadium center in charge storage processes. Additionally, the custom flexible sensing patch presents an excellent sensitivity retention rate under bending (95%) and twisting (93%) strains and possesses good anti-interference performance (ΔE < 8 mV) against common interfering ions and organic substances in sweat. Real-time monitoring of the Cl- concentration in sweat aligns with ion chromatography analysis results. This study presents a compact wearable Cl- monitoring platform for the easy tracking of exercise-induced dehydration and cystic fibrosis screening with promising applications in smart healthcare.

PMID:40014548 | DOI:10.1021/acssensors.4c03366

Proc Natl Acad Sci U S A. 2025 Mar 4;122(9):e2418643122. doi: 10.1073/pnas.2418643122. Epub 2025 Feb 27.

ABSTRACT

Single-particle tracking has enabled quantitative studies of complex systems, providing nanometer localization precision and millisecond temporal resolution in heterogeneous environments. However, at micro- or nanometer scales, probe dynamics become inherently stochastic due to Brownian motion and complex interactions, leading to varied diffusion behaviors. Typically, analysis of such trajectory data involves certain moving-window operation and assumes the existence of some pseudo-steady states, particularly when evaluating predefined parameters or specific types of diffusion modes. Here, we introduce the stochastic particle-informed neural network (SPINN), a physics-informed deep learning framework that integrates stochastic differential equations to model and infer particle diffusion dynamics. The SPINN autonomously explores parameter spaces and distinguishes between deterministic and stochastic components with single-frame resolution. Using the anomalous diffusion dataset, we validated SPINN's ability to reduce frame-to-frame variability while preserving key statistical correlations, allowing for accurate characterization of different stochastic processes. When applied to the diffusion of single gold nanorods in hydrogels, the SPINN revealed enhanced microrheological properties during hydrogel gelation and uncovered interfacial dynamics during dextran/tetra-PEG liquid-liquid phase separation. By improving the temporal resolution of stochastic dynamics, the SPINN facilitates the estimation and prediction of complex diffusion behaviors, offering insights into underlying physical mechanisms at mesoscopic scales.

PMID:40014572 | DOI:10.1073/pnas.2418643122

Microsc Res Tech. 2025 Feb 27. doi: 10.1002/jemt.24836. Online ahead of print.

ABSTRACT

The retina is an important part of the eye that aids in focusing light and visual recognition to the brain. Hence, its damage causes vision loss in the human eye. Central serous retinopathy is a common retinal disorder in which serous detachment occurs at the posterior pole of the retina. Therefore, detection of CSR at an early stage with good accuracy can decrease the rate of vision loss and recover the vision to normal conditions. In the past, numerous manual techniques have been devised for CSR detection; nevertheless, they have demonstrated imprecision and unreliability. Thus, the deep learning method can play an important role in automatically detecting CSR. This research presents a convolutional neural network-based framework combined with segmentation and post-ocessing for CSR classification. There are several challenges in the segmentation of retinal images, such as noise, size variation, location, and shape of the fluid in the retina. To address these limitations, Otsu's thresholding has been employed as a technique for segmenting optical coherence tomography (OCT) images. Pigments and fluids are present in epithelial detachment, and contrast adjustment and noise removal are required. After segmentation, post-processing is used, combining flood filling, dilation, and area thresholding. The segmented processed OCT scans were classified using the fusion of three networks: (i) ResNet-18, (ii) Google-Net, and (iii) VGG-19. After experimentation, the fusion of ResNet-18, GoogleNet, and VGG-19 achieved 99.6% accuracy, 99.46% sensitivity, 100% specificity, and 99.73% F1 score using the proposed framework for classifying normal and CSR-affected images. A publicly available dataset OCTID comprises 207 normal and 102 CSR-affected images was utilized for testing and training of the proposed method. The experimental findings conclusively demonstrate the inherent suitability and efficacy of the framework put forth. Through rigorous testing and analysis, the results unequivocally validate the framework's ability to fulfill its intended objectives and address the challenges at hand.

PMID:40014549 | DOI:10.1002/jemt.24836

Magn Reson Med. 2025 Feb 27. doi: 10.1002/mrm.30478. Online ahead of print.

ABSTRACT

PURPOSE: The clinical value of quantitative MRI hinges on its measurement repeatability. Deep learning methods to reconstruct undersampled quantitative MRI can accelerate reconstruction but do not aim to promote quantitative repeatability. This study proposes a repeatability-encouraging self-supervised learning (SSL) reconstruction method for quantitative MRI.

METHODS: The proposed SSL reconstruction network minimized cross-data-consistency between two equally sized, mutually exclusive temporal subsets of k-t-space data, encouraging repeatability by enabling each subset's reconstruction to predict the other's k-t-space data. The method was evaluated on cardiac MR Multitasking T1 mapping data and compared with supervised learning methods trained on full 60-s inputs (Sup60) and on split 30-s inputs (Sup30/30). Reconstruction quality was evaluated on full 60-s inputs, comparing results to iterative wavelet-regularized references using Bland-Altman limits of agreement (LOAs). Repeatability was evaluated by splitting the 60-s data into two 30-s inputs, evaluating T1 differences between reconstructions from the two halves of the scan.

RESULTS: On 60-s inputs, the proposed method produced comparable-quality images and T1 maps to the Sup60 method, with T1 values in general agreement with the wavelet reference (LOA Sup60 = ±75 ms, SSL = ±81 ms), whereas the Sup30/30 method generated blurrier results and showed poor T1 agreement (LOA Sup30/30 = ±132 ms). On back-to-back 30-s inputs, the proposed method had the best T1 repeatability (coefficient of variation SSL = 6.3%, Sup60 = 12.0%, Sup30/30 = 6.9%). Of the three deep learning methods, only the SSL method produced sharp and repeatable images.

CONCLUSION: Without the need for labeled training data, the proposed SSL method demonstrated superior repeatability compared with supervised learning without sacrificing sharpness, and reduced reconstruction time versus iterative methods.

PMID:40014485 | DOI:10.1002/mrm.30478

ACS Appl Bio Mater. 2025 Feb 27. doi: 10.1021/acsabm.4c01621. Online ahead of print.

ABSTRACT

Bacterial microcompartments (BMCs) are self-assembling protein shell structures that are widely investigated across a broad range of biological and abiotic chemistry applications. A central challenge in BMC research is the targeted capture of enzymes during shell assembly. While crystallography and cryo-EM techniques have been successful in determining BMC shell structures, there has been only limited success in visualizing the location of BMC-captured enzyme cargo. Here, we demonstrate the opportunity to use small-angle X-ray scattering (SAXS) and pair distance distribution function (PDDF) measurements combined with quantitative comparison to coordinate structure models as an approach to characterize BMC shell structures in solution conditions directly relevant to biochemical function. Using this approach, we analyzed BMC shells from Haliangium ochraceum (HO) that were isolated following expression in E. coli. The analysis allowed the BMC shell structures and the extent of encapsulated enzyme cargo to be identified. Notably, the results demonstrate that HO-BMC shells adventitiously capture significant amounts of cytoplasmic cargo during assembly in E. coli. Our findings highlight the utility of SAXS/PDDF analysis for evaluating BMC architectures and enzyme encapsulation, offering valuable insights for designing BMC shells as platforms for biological and abiotic catalyst capture within confined environments.

PMID:40014870 | DOI:10.1021/acsabm.4c01621

ACS Nano. 2025 Feb 27. doi: 10.1021/acsnano.4c16988. Online ahead of print.

ABSTRACT

Exosomes present in the circulatory system demonstrate considerable promise for the diagnosis and treatment of diseases. Nevertheless, the complex nature of blood samples and the prevalence of highly abundant proteins pose a significant obstacle to prompt and effective isolation and functional evaluation of exosomes from blood. Here, we present a fully integrated lab-on-a-disc equipped with two nanofilters, also termed iExoDisc, which facilitates automated isolation of exosomes from 400 μL blood samples within 45 min. By integrating the plasma separation module, highly abundant protein removal module, and nanopore membrane-based total isolation module, the resulting exosomes exhibited significantly increased purity (∼3-6-fold) compared to conventional ultracentrifugation and polymer precipitation. Additionally, we then successfully performed nontargeted and targeted glycan profiling on exosomes derived from clinical triple-negative breast cancer (TNBC) patients using MALDI-TOF-MS and lectin microarray containing 56 kinds of lectins. The findings from both methodologies indicated that galactosylation and sialylation exhibit potential as diagnostic indicators for TNBC. Finally, by utilizing the exosome-specific glycosylated protein CD63 as a proof-of-concept, we successfully realized the integration of point-of-care on-chip exosome separation and in situ detection with 2 h. Thus, the iExoDisc provides a potential approach to early cancer detection, liquid biopsy, and point-of-care diagnosis.

PMID:40014808 | DOI:10.1021/acsnano.4c16988

Science. 2025 Feb 27:eadv9789. doi: 10.1126/science.adv9789. Online ahead of print.

ABSTRACT

RNA-guided systems provide remarkable versatility, enabling diverse biological functions. Through iterative structural and sequence homology-based mining starting with a guide RNA-interaction domain of Cas9, we identified a family of RNA-guided DNA-targeting proteins in phage and parasitic bacteria. Each system consists of a Tandem Interspaced Guide RNA (TIGR) array and a TIGR-associated (Tas) protein containing a Nop domain, sometimes fused to HNH (TasH) or RuvC (TasR) nuclease domains. We show that TIGR arrays are processed into 36-nt RNAs (tigRNAs) that direct sequence-specific DNA binding through a tandem-spacer targeting mechanism. TasR can be reprogrammed for precise DNA cleavage, including in human cells. The structure of TasR reveals striking similarities to box C/D snoRNPs and IS110 RNA-guided transposases, providing insights into the evolution of diverse RNA-guided systems.

PMID:40014690 | DOI:10.1126/science.adv9789

Science. 2025 Feb 27:eadk3248. doi: 10.1126/science.adk3248. Online ahead of print.

ABSTRACT

During infections, CD4+ Foxp3+ regulatory T (Treg) cells must control autoreactive CD4+ conventional T (Tconv) cell responses against self-peptide antigens while permitting those against pathogen-derived "nonself" peptides. We defined the basis of this selectivity using mice in which Treg cells reactive to a single prostate-specific self-peptide were selectively depleted. We found that self-peptide-specific Treg cells were dispensable for the control of Tconv cells of matched specificity at homeostasis. However, they were required to control such Tconv cells and prevent autoimmunity toward the prostate following exposure to elevated self-peptide during infection. Importantly, the Treg cell response to self-peptide did not impact protective Tconv cell responses to a pathogen-derived peptide. Thus, self-peptide-specific Treg cells promoted self-nonself discrimination during infection by selectively controlling Tconv cells of shared self-specificity.

PMID:40014689 | DOI:10.1126/science.adk3248

Angew Chem Int Ed Engl. 2025 Feb 27:e202419510. doi: 10.1002/anie.202419510. Online ahead of print.

ABSTRACT

Understanding root-bacteria interactions with plant growth-promoting rhizobacteria (PGPR) is key to developing effective biofertilizers for sustainable agriculture. We performed single-cell force spectroscopy using the atomic force microscope (AFM) to study the primary attachment of two PGPR, Bacillus velezensis and Pseudomonas defensor, to different regions of Arabidopsis thaliana roots. Force measurements with individual cells uncovered distinct attachment strategies by each strain, involving binding via micrometer-long polymers from both bacteria and root surfaces. Flagella differentially affected the binding interactions of each PGPR; their removal altered binding characteristics differently for each strain, highlighting the importance of surface polymeric molecules in early root colonization. Using silica beads to mimic the negatively charged bacteria, we demonstrated the influence of electrostatic forces on root-bacteria interactions. We examined interactions with abiotic surfaces of varying surface energies, revealing the roles of hydrophilic and hydrophobic forces in initial binding. Our measurements show that differences in physico-chemical properties of bacteria and roots are responsible for variations in primary attachment strategies between PGPR strains and root regions. Parallel fluorescence measurements corroborated our AFM single-cell analysis. Overall, our results provide a nanoscale view of bacterial attachment to roots, offering key insights into how beneficial bacteria colonize roots, crucial for enhancing biofertilizer effectiveness.

PMID:40014612 | DOI:10.1002/anie.202419510

Cell Rep. 2025 Feb 26;44(3):115357. doi: 10.1016/j.celrep.2025.115357. Online ahead of print.

ABSTRACT

Despite advances in cancer treatment, the development of effective therapies remains an urgent unmet need. Here, we investigate the potential of bacteria-derived metabolites as a therapeutic alternative for the treatment of cancer. We detect 3-hydroxydodecanedioic acid in the serum of tumor-bearing mice treated with serum from mice previously supplemented with a mix of Clostridiales bacteria. Further, 3-hydroxydodecanoic acid, an intermediate derivative between dodecanoic and 3-hydroxydodecanedioic acids, exhibits a strong anti-tumor response via GPR84 receptor signaling and enhances CD8+ T cell infiltration and cytotoxicity within tumor tissue in multiple cancer models. Metabolomics analysis of colorectal cancer patient serum reveals an inverse correlation between the abundance of these metabolites and advanced disease stages. Our findings provide a strong rationale for 3-hydroxydodecanoic acid and the GPR84 receptor to be considered as promising therapeutic targets for cancer treatment.

PMID:40014452 | DOI:10.1016/j.celrep.2025.115357

J Clin Psychopharmacol. 2025 Mar-Apr 01;45(2):111-115. doi: 10.1097/JCP.0000000000001962.

ABSTRACT

PURPOSE: Women have historically been underrepresented in second-generation antipsychotic (SGA) clinical trials, accounting for less than 35% of participants, which raises concerns about the generalizability of the safety profile for these medications.

METHODS: The US adverse event reporting system was queried for the dates January 1, 2019, to July 8, 2024, to examine the following 6 SGAs: aripiprazole, clozapine, olanzapine, quetiapine, risperidone, and ziprasidone. Reports were excluded if patients were under 18 years old, contained an unknown age or gender, or were duplicated. Five adverse events were examined: Torsades de pointes (TdP), neuroleptic malignant syndrome (NMS), tardive dyskinesia (TD), agranulocytosis (AG), and cerebrovascular adverse events (CVAE). Counts of these events were noted, and reporting odds ratios (ROR) were calculated.

RESULTS: The total study cohort was 87,356 reports, consisting of aripiprazole (n = 10,715, 12.2%), clozapine (n = 25,096, 28.7%), olanzapine (n = 11,587, 13.3%), quetiapine (n = 28,746, 32.9%), risperidone (n = 10,467, 12%), and ziprasidone (n = 745, 0.9%). The cohort's mean age was 48.6 ± 18.5 years and comprised 42,584 females (48.7%). Most cases were reported by healthcare professionals (74,836, 85.7%). A total of 3,754 reports contained at least 1 of the 5 adverse events. The RORs among females compared to males for TdP (5.55, 95% confidence interval [CI] = 3.78-8.47), NMS (0.59, 95% CI = 0.53-0.65), TD (0.88, 95% CI = 0.76-1.02), AG (0.59, 95% CI = 0.51-0.70), and CVAE (1.12, 95% CI = 0.89-1.41) were observed. Females had a significantly higher odds of hospitalization or death with TdP compared to males (ROR = 3.09, 95% CI = 1.36-7.01).

CONCLUSIONS: Our findings suggest higher odds of TdP and worse TdP-associated outcomes among females exposed to SGAs compared to males. Further studies are needed to confirm these preliminary findings.

PMID:40014466 | DOI:10.1097/JCP.0000000000001962

Turk J Ophthalmol. 2025 Feb 27;55(1):49-52. doi: 10.4274/tjo.galenos.2024.23934.

ABSTRACT

The purpose of this report is to present simultaneous bilateral penetrating keratoplasty (PK) in Acanthamoeba keratitis (AK). A 42-year-old male with keratoconus, wearing bilateral hybrid contact lenses, presented with pain in the left eye. He had a history of intrastromal corneal ring segment placement in the right and PK in the left eye. His best corrected visual acuity (BCVA) was 20/640 in the right eye and 20/2000 in the left. Slit-lamp examination revealed a ring-shaped infiltration on the left. Despite two months of broad-spectrum topical antibiotic therapy, microbiological examination of corneal scraping samples was repeated but revealed no evidence of microbial agents. In vivo confocal microscopy findings were not compatible with AK. During the follow-up, corneal infiltration and stromal melt were observed in the right eye. Corneal scraping samples from the right eye were sent for microbiological examination, but again no microbial agents were identified. Histopathological examination revealed spherical cysts consistent with AK. Corneal perforation developed in the right eye, while simultaneous wound dehiscence occurred in the left eye. Since the patient had a history of renal failure, simultaneous bilateral tectonic-therapeutic PK was performed to minimize the risks arising from general anesthesia. Postoperative BCVA was 20/50 in the right eye and 20/125 in the left eye at 6 months. Diagnostic tools can be misleading in eyes with altered anatomy. Careful examination and a timely decision to perform tectonic-therapeutic PK are vital in preventing devastating complications.

PMID:40013506 | DOI:10.4274/tjo.galenos.2024.23934

Pediatr Nephrol. 2025 Feb 27. doi: 10.1007/s00467-025-06728-y. Online ahead of print.

ABSTRACT

BACKGROUND: Pediatric palliative care (PPC) aims to improve the quality of life for children with life-limiting conditions, such as advanced chronic kidney disease (CKD), from the time of diagnosis. However, PPC is not commonly integrated into routine pediatric nephrology care. This study explores the perspectives and experiences of healthcare providers (HCPs) to better understand the experiences and specific barriers related to PPC integration for children and adolescents with advanced CKD.

METHODS: We conducted a qualitative study with 23 HCPs, including nurses, psychologists, social workers, and physicians from seven German pediatric nephrology centers, analyzing semi-structured focus groups and individual interviews using structured content analysis.

RESULTS: Five main categories emerged from the analysis, revealing HCPs' perceptions of CKD as a life-limiting condition, HCPs' moral distress in addressing end-of-life issues, and barriers to PPC integration. Although HCPs reported comprehensive multidisciplinary support for end-of-life situations, a lack of interprofessional communication occasionally hindered coordinated care. HCPs rarely addressed CKD's life-limiting nature proactively. A fear of diminishing hope led HCPs to avoid conversations about prognosis unless in response to a therapeutic crisis. PPC was mostly reserved for end-of-life cases, as HCPs associated PPC with terminal care and expressed concerns over distressing families.

CONCLUSIONS: This study highlights the gap between guidelines recommending early integration of PPC and daily nephrology practice, which tends to introduce PPC late in the course of the disease. Training for nephrology teams could improve the quality of life for children with advanced CKD and their families by promoting early integration of primary PPC principles.

PMID:40014135 | DOI:10.1007/s00467-025-06728-y

Respir Care. 2025 Feb 10. doi: 10.1089/respcare.12352. Online ahead of print.

ABSTRACT

Asthma experienced by both adults and children is a phenotypically heterogeneous condition. Severe asthma, characterized by ongoing symptoms and airway inflammation despite high doses of inhaled and/or systemic corticosteroids, is the focus of research efforts to understand this underlying heterogeneity. Clinical phenotypes in both adult and pediatric asthma have been determined using supervised definition-driven classification and unsupervised data-driven clustering methods. Efforts to understand the underlying inflammatory patterns of severe asthma have led to the seminal discovery of type 2-high versus type 2-low phenotypes and to the development of biologics targeted at type 2-high inflammation to reduce the rates of severe asthma exacerbations. Type 2-high asthma is characterized by upregulation of T helper 2 immune pathways including interleukin (IL)-4, IL-5, and IL-13 along with eosinophilic airway inflammation, sometimes allergic sensitization, and responsiveness to treatment with corticosteroids. Type 2-low asthma is poorly responsive to corticosteroids and is not as well characterized as type 2-high asthma. Type 2-low asthma is limited by being defined as the absence of type 2-high inflammatory markers. Choosing a biologic for the treatment of severe asthma involves the evaluation of a panel of biomarkers such as blood eosinophils, total and specific immunoglobulin E/allergic sensitization, and fractional exhaled nitric oxide. In this review, we focus on the underlying pathobiology of adult and pediatric asthma, discuss the different phenotype-based treatment options for adult and pediatric type 2-high with or without allergic asthma and type 2-low asthma, and describe a clinical phenotyping approach to patients to guide out-patient therapy. Finally, we end with a discussion of whether pediatric asthma exacerbations necessitating admission to an ICU constitute their own high-risk phenotype and/or whether it is a part of other previously defined high-risk subgroups such as difficult-to-control asthma, exacerbation-prone asthma, and severe treatment-resistant asthma.

PMID:40013975 | DOI:10.1089/respcare.12352

Lung India. 2025 Mar 1;42(2):103-108. doi: 10.4103/lungindia.lungindia_404_24. Epub 2025 Feb 27.

ABSTRACT

BACKGROUND: Cystic fibrosis (CF) is a genetic disorder caused by genetic variant in the cystic fibrosis transmembrane regulator (CFTR) gene that affects around 89,000 people worldwide. Loss of the CFTR chloride channel due to pathogenic variants in the CFTR gene causes obstruction in the exocrine pancreas gland and reduced lung function.

OBJECTIVE: To determine the genotype and phenotype of patients with CF from western India.

MATERIALS AND METHODS: This was a single-center retrospective cross-sectional study conducted in a tertiary care super speciality paediatric hospital of Mumbai, India, comprising patients aged 0 to 18 years visiting a paediatric pulmonology clinic with suspected or confirmed diagnosis of CF.

RESULTS: The mean (SD) age of onset of symptoms was 6.8 (10.2) months and the mean (SD) age at diagnosis was 32.5 (50.5) months. The two most common genetic variants found in our patients were c. 1521_1523delCTT (F508del) (n = 21) and c.1367T>C (V456A) (n = 10). There were nine novel genetic variants identified that have not been reported so far. The mean (SD) age of onset of symptoms was 6.8 (10.2) months and mean (SD) age at diagnosis was 32.5 (50.5) months. The most common presenting features were recurrent respiratory infections (83%), malabsorption (79%), and failure to thrive (79%). Sweat chloride testing was conducted to establish the CFTR gene dysfunction and was positive in 79% (46/58) of patients and intermediate in 15% (n = 9/58) of patients. The two most common genetic variants found in our group of patients were c. 1521_1523delCTT (F508del) (n = 21) and c.1367T>C (V456A) (n = 10). There were nine novel genetic variants identified that have not been reported so far.

CONCLUSION: This study adds to the knowledge of genetic diversity in the pathogenic CFTR gene variants causing CF and highlights the importance of sequencing the entire CFTR gene as regional variations in the gene have been documented in India.

PMID:40013628 | DOI:10.4103/lungindia.lungindia_404_24

Transl Vis Sci Technol. 2025 Feb 3;14(2):26. doi: 10.1167/tvst.14.2.26.

ABSTRACT

PURPOSE: This study aims to develop an automated pipeline to detect retinal detachment from B-scan ocular ultrasonography (USG) images by using deep learning-based segmentation.

METHODS: A computational pipeline consisting of an encoder-decoder segmentation network and a machine learning classifier was developed, trained, and validated using 279 B-scan ocular USG images from 204 patients, including 66 retinal detachment (RD) images, 36 posterior vitreous detachment images, and 177 healthy control images. Performance metrics, including the precision, recall, and F-scores, were calculated for both segmentation and RD detection.

RESULTS: The overall pipeline achieved 96.3% F-score for RD detection, outperforming end-to-end deep learning classification models (ResNet-50 and MobileNetV3) with 94.3% and 95.0% F-scores. This improvement was also validated on an independent test set, where the proposed pipeline led to 96.5% F-score, but the classification models yielded only 62.1% and 84.9% F-scores, respectively. Besides, the segmentation model of this pipeline led to high performances across multiple ocular structures, with 84.7%, 78.3%, and 88.2% F-scores for retina/choroid, sclera, and optic nerve sheath segmentation, respectively. The segmentation model outperforms the standard UNet, particularly in challenging RD cases, where it effectively segmented detached retina regions.

CONCLUSIONS: The proposed automated segmentation and classification method improves RD detection in B-scan ocular USG images compared to end-to-end classification models, offering potential clinical benefits in resource-limited settings.

TRANSLATIONAL RELEVANCE: We have developed a novel deep/machine learning based pipeline that has the potential to significantly improve diagnostic accuracy and accessibility for ocular USG.

PMID:40014336 | DOI:10.1167/tvst.14.2.26

Eur J Nucl Med Mol Imaging. 2025 Feb 27. doi: 10.1007/s00259-025-07149-7. Online ahead of print.

ABSTRACT

PURPOSE: PET/CT imaging data contains a wealth of quantitative information that can provide valuable contributions to characterising tumours. A growing body of work focuses on the use of deep-learning (DL) techniques for denoising PET data. These models are clinically evaluated prior to use, however, quantitative image assessment provides potential for further evaluation. This work uses radiomic features to compare two manufacturer deep-learning (DL) image enhancement algorithms, one of which has been commercialised, against 'gold-standard' image reconstruction techniques in phantom data and a sarcoma patient data set (N=20).

METHODS: All studies in the retrospective sarcoma clinical [ 18 F]FDG dataset were acquired on either a GE Discovery 690 or 710 PET/CT scanner with volumes segmented by an experienced nuclear medicine radiologist. The modular heterogeneous imaging phantom used in this work was filled with [ 18 F]FDG, and five repeat acquisitions of the phantom were acquired on a GE Discovery 710 PET/CT scanner. The DL-enhanced images were compared to 'gold-standard' images the algorithms were trained to emulate and input images. The difference between image sets was tested for significance in 93 international biomarker standardisation initiative (IBSI) standardised radiomic features.

RESULTS: Comparing DL-enhanced images to the 'gold-standard', 4.0% and 9.7% radiomic features measured significantly different (pcritical < 0.0005) in the phantom and patient data respectively (averaged over the two DL algorithms). Larger differences were observed comparing DL-enhanced images to algorithm input images with 29.8% and 43.0% of radiomic features measuring significantly different in the phantom and patient data respectively (averaged over the two DL algorithms).

CONCLUSION: DL-enhanced images were found to be similar to images generated using the 'gold-standard' target image reconstruction method with more than 80% of radiomic features not significantly different in all comparisons across unseen phantom and sarcoma patient data. This result offers insight into the performance of the DL algorithms, and demonstrate potential applications for DL algorithms in harmonisation for radiomics and for radiomic features in quantitative evaluation of DL algorithms.

PMID:40014074 | DOI:10.1007/s00259-025-07149-7

J Appl Clin Med Phys. 2025 Feb 27:e70034. doi: 10.1002/acm2.70034. Online ahead of print.

ABSTRACT

Artificial intelligence (AI) has the potential to revolutionize brachytherapy's clinical workflow. This review comprehensively examines the application of AI, focusing on machine learning and deep learning, in various aspects of brachytherapy. We analyze AI's role in making brachytherapy treatments more personalized, efficient, and effective. The applications are systematically categorized into seven categories: imaging, preplanning, treatment planning, applicator reconstruction, quality assurance, outcome prediction, and real-time monitoring. Each major category is further subdivided based on cancer type or specific tasks, with detailed summaries of models, data sizes, and results presented in corresponding tables. Additionally, we discuss the limitations, challenges, and ethical concerns of current AI applications, along with perspectives on future directions. This review offers insights into the current advancements, challenges, and the impact of AI on treatment paradigms, encouraging further research to expand its clinical utility.

PMID:40014044 | DOI:10.1002/acm2.70034

J Neurophysiol. 2025 Feb 27. doi: 10.1152/jn.00622.2024. Online ahead of print.

ABSTRACT

Dendritic spines, small protrusions on neuronal dendrites, play a crucial role in brain function by changing shape and size in response to neural activity. So far, in depth analysis of dendritic spines in human brain tissue is lacking. This study presents a comprehensive analysis of human dendritic spine morphology and density using a unique dataset from human brain tissue from 27 patients (8 females, 19 males, aged 18-71) undergoing tumor or epilepsy surgery at three neurosurgery sites. We used acute slices and organotypic brain slice cultures to examine dendritic spines, classifying them into the three main morphological subtypes: Mushroom, Thin, and Stubby, via 3D reconstruction using ZEISS arivis Pro software. A deep learning model, trained on 39 diverse datasets, automated spine segmentation and 3D reconstruction, achieving a 74% F1-score and reducing processing time by over 50%. We show significant differences in spine density by sex, dendrite type, and tissue condition. Females had higher spine densities than males, and apical dendrites were denser in spines than basal ones. Acute tissue showed higher spine densities compared to cultured human brain tissue. With time in culture, Mushroom spines decreased, while Stubby and Thin spine percentages increased, particularly from 7-9 to 14 days in vitro, reflecting potential synaptic plasticity changes. Our study underscores the importance of using human brain tissue to understand unique synaptic properties and shows that integrating deep learning with traditional methods enables efficient large-scale analysis, revealing key insights into sex- and tissue-specific dendritic spine dynamics relevant to neurological diseases.

PMID:40013734 | DOI:10.1152/jn.00622.2024

J Chem Inf Model. 2025 Feb 27. doi: 10.1021/acs.jcim.4c02051. Online ahead of print.

ABSTRACT

Cytochrome P450 enzymes (CYP450s) play crucial roles in metabolizing many drugs, and thus, local chemical structure can profoundly influence drug efficacy and toxicity. Therefore, the accurate prediction of CYP450-mediated reaction sites can increase the efficiency of drug discovery and development. Here, we present GLMCyp, a deep learning-based approach, for predicting CYP450 reaction sites on small molecules. By integrating two-dimensional (2D) molecular graph features, three-dimensional (3D) features from Uni-Mol, and relevant CYP450 protein features generated by ESM-2, GLMCyp could accurately predict bonds of metabolism (BoMs) targeted by a panel of nine human CYP450s. Incorporating protein features allowed GLMCyp application in broader CYP450 metabolism prediction tasks. Additionally, substrate molecular feature processing enhanced the accuracy and interpretability of the predictions. The model was trained on the EBoMD data set and reached an area under the receiver operating characteristic curve (ROC-AUC) of 0.926. GLMCyp also showed a relatively strong capacity for feature extraction and generalizability in validation with external data sets. The GLMCyp model and data sets are available for public use (https://github.com/lvimmind/GLMCyp-Predictor) to facilitate drug metabolism screening.

PMID:40013456 | DOI:10.1021/acs.jcim.4c02051