JMIR Diabetes. 2025 Apr 18;10:e75497. doi: 10.2196/75497.

NO ABSTRACT

PMID:40249833 | DOI:10.2196/75497

IEEE Trans Image Process. 2025 Apr 18;PP. doi: 10.1109/TIP.2025.3558425. Online ahead of print.

ABSTRACT

Semantic segmentation is an important branch of image processing and computer vision. With the popularity of deep learning, various convolutional neural networks have been proposed for pixel-level classification and segmentation tasks. In practical scenarios, however, imaging angles are often arbitrary, encompassing instances such as water body images from remote sensing and capillary and polyp images in the medical domain, where prior orientation information is typically unavailable to guide these networks to extract more effective features. In this case, learning features from objects with diverse orientation information poses a significant challenge, as the majority of CNN-based semantic segmentation networks lack rotation equivariance to resist the disturbance from orientation information. To address this challenge, this paper first constructs a universal convolutiongroup framework aimed at more fully utilizing orientation information and equipping the network with rotation equivariance. Subsequently, we mathematically design a padding-based rotation equivariant convolution mode (PreCM), which is not only applicable to multi-scale images and convolutional kernels but can also serve as a replacement component for various types of convolutions, such as dilated convolutions, transposed convolutions, and asymmetric convolution. To quantitatively assess the impact of image rotation in semantic segmentation tasks, we also propose a new evaluation metric, Rotation Difference (RD). The replacement experiments related to six existing semantic segmentation networks on three datasets (i.e., Satellite Images of Water Bodies, DRIVE, and Floodnet) show that, the average Intersection Over Union (IOU) of their PreCM-based versions respectively improve 6.91%, 10.63%, 4.53%, 5.93%, 7.48%, 8.33% compared to their original versions in terms of random angle rotation. And the average RD values are decreased by 3.58%, 4.56%, 3.47%, 3.66%, 3.47%, 3.43% respectively. The code can be download from https://github.com/XinyuXu414.

PMID:40249694 | DOI:10.1109/TIP.2025.3558425

IEEE Trans Image Process. 2025 Apr 18;PP. doi: 10.1109/TIP.2025.3560430. Online ahead of print.

ABSTRACT

Snapshot Spectral Imaging (SSI) techniques, with the ability to capture both spectral and spatial information in a single exposure, have been found useful in a wide range of applications. SSI systems generally operate within the 'encoding-decoding' framework, leveraging the synergism of optical hardware and reconstruction algorithms. Typically, reconstructing desired spectral images from SSI measurements is an ill-posed and challenging problem. Existing studies utilize either model-based or deep learning-based methods, but both have their drawbacks. Model-based algorithms suffer from high computational costs, while supervised learning-based methods rely on large paired training data. In this paper, we propose a novel Unsupervised range-Nullspace learning (UnNull) prior for spectral image reconstruction. UnNull explicitly models the data via subspace decomposition, offering enhanced interpretability and generalization ability. Specifically, UnNull considers that the spectral images can be decomposed into the range and null subspaces. The features projected onto the range subspace are mainly low-frequency information, while features in the nullspace represent high-frequency information. Comprehensive multispectral demosaicing and reconstruction experiments demonstrate the superior performance of our proposed algorithm.

PMID:40249693 | DOI:10.1109/TIP.2025.3560430

IEEE Trans Med Imaging. 2025 Apr 18;PP. doi: 10.1109/TMI.2025.3562443. Online ahead of print.

ABSTRACT

Mesh-based cortical surface reconstruction is essential for neuroimaging, enabling precise measurements of brain morphology such as cortical thickness. Establishing vertex correspondence between individual cortical meshes and group templates allows vertex-level comparisons, but traditional methods require time-consuming post-processing steps to achieve vertex correspondence. While deep learning has improved accuracy in cortical surface reconstruction, optimizing vertex correspondence has not been the focus of prior work. We introduce Vox2Cortex with Correspondence (V2CC), an extension of Vox2Cortex, which replaces the commonly used Chamfer loss with L1 loss on registered surfaces. This approach improves inter- and intra-subject correspondence, which makes it suitable for direct group comparisons and atlas-based parcellation. Additionally, we analyze mesh self-intersections, categorizing them into minor (neighboring faces) and major (non-neighboring faces) types.To address major self-intersections, which are not effectively handled by standard regularization losses, we propose a novel Self-Proximity loss, designed to adjust non-neighboring vertices within a defined proximity threshold. Comprehensive evaluations demonstrate that recent deep learning methods inadequately address vertex correspondence, often causing in-accuracies in parcellation. In contrast, our method achieves accurate correspondence and reduces self-intersections to below 1% for both pial and white matter surfaces.

PMID:40249681 | DOI:10.1109/TMI.2025.3562443

Sci Immunol. 2025 Apr 18;10(106):eadr4977. doi: 10.1126/sciimmunol.adr4977. Epub 2025 Apr 18.

ABSTRACT

Lung interstitial macrophages (IMs) are monocyte-derived parenchymal macrophages whose tissue-supportive functions remain unclear. Despite progress in understanding lung IM diversity and transcriptional regulation, the signals driving their development from monocytes and their functional specification remain unknown. Here, we found that lung endothelial cell-derived Tgfβ1 triggered a core Tgfβ receptor-dependent IM signature in mouse bone marrow-derived monocytes. Myeloid-specific impairment of Tgfβ receptor signaling severely disrupted monocyte-to-IM development, leading to the accumulation of perivascular immature monocytes, reduced IM numbers, and a loss of IM-intrinsic identity, a phenomenon similarly observed in the absence of endothelial-specific Tgfβ1. Mice lacking the Tgfβ receptor in monocytes and IMs exhibited altered monocyte and IM niche occupancy and hallmarks of aging including impaired immunoregulation, hyperinflation, and fibrosis. Our work identifies a Tgfβ signaling-dependent endothelial-IM axis that shapes IM development and sustains lung integrity, providing foundations for IM-targeted interventions in aging and chronic inflammation.

PMID:40249827 | DOI:10.1126/sciimmunol.adr4977

Sci Adv. 2025 Apr 18;11(16):eads4609. doi: 10.1126/sciadv.ads4609. Epub 2025 Apr 18.

ABSTRACT

Molecular movement and interactions at the single-molecule level, particularly in live cells, are often studied using fluorescence correlation spectroscopy (FCS). While powerful, FCS has notable drawbacks: It requires high laser intensities and long acquisition times, increasing phototoxicity, and often relies on problematic statistical assumptions in data fitting. We introduce fluorescence intensity trace statistical analysis (FITSA), a Bayesian method that directly analyzes fluorescence intensity traces. FITSA offers faster, more stable convergence than previous approaches and provides robust parameter estimation from far shorter measurements than conventional FCS. Our results demonstrate that FITSA achieves comparable precision to FCS while requiring substantially fewer photons. This advantage becomes even more pronounced when accounting for statistical dependencies in FCS analysis, which are often overlooked but necessary for accurate error estimation. By reducing laser exposure, FITSA minimizes phototoxicity effects, representing a major advancement in the quantitative analysis of molecular processes across fields.

PMID:40249821 | DOI:10.1126/sciadv.ads4609

Proc Natl Acad Sci U S A. 2025 Apr 22;122(16):e2424474122. doi: 10.1073/pnas.2424474122. Epub 2025 Apr 18.

ABSTRACT

Bestrophin-1 (BEST1) is a chloride channel expressed in the eye and other tissues of the body. A link between BEST1 and the principal inhibitory neurotransmitter γ-aminobutyric acid (GABA) has been proposed. The most appreciated receptors for extracellular GABA are the GABAB G-protein-coupled receptors and the pentameric GABAA chloride channels, both of which have fundamental roles in the central nervous system. Here, we demonstrate that BEST1 is directly activated by GABA. Through functional studies and atomic-resolution structures of human and chicken BEST1, we identify a GABA binding site on the channel's extracellular side and determine the mechanism by which GABA binding stabilizes opening of the channel's central gate. This same gate, "the neck," is activated by intracellular [Ca2+], indicating that BEST1 is controlled by ligands from both sides of the membrane. The studies demonstrate that BEST1, which shares no structural homology with GABAA receptors, is a GABA-activated chloride channel. The physiological implications of this finding remain to be studied.

PMID:40249777 | DOI:10.1073/pnas.2424474122

Cancer Med. 2025 Apr;14(8):e70853. doi: 10.1002/cam4.70853.

ABSTRACT

BACKGROUND: Genetic testing and sequencing technologies offer a comprehensive understanding of cancer genetics, providing rapid and cost-effective solutions. In particular, these advanced technologies play an important role in assessing the complexities of the rare cancer types affecting several systems including the bone, endocrine, digestive, vascular, and soft tissue. This review will explore how genetic testing and sequencing technologies have contributed to the identification of biomarkers across several rare cancer types in diagnostic, therapeutic, and prognostic stages, thereby advancing PM.

METHODS: A comprehensive literature search was conducted across PubMed (MEDLINE), EMBASE, and Web of Science using keywords related to sequencing technologies, genetic testing, and cancer. There were no restrictions on language, methodology, age, or publication date. Both primary and secondary research involving humans or animals were considered.

RESULTS: In practice, fluorescence in situ hybridization, karyotype, microarrays and other genetic tests are mainly applied to identify specific genetic alterations and mutations associated with cancer progression. Sequencing technologies, such as next generation sequencing, polymerase chain reaction, whole genome or exome sequencing, enable the rapid analysis of millions of DNA fragments. These techniques assess genome structure, genetic changes, gene expression profiles, and epigenetic variations. Consequently, they help detect main intrinsic markers that are crucial for personalizing diagnosis, treatment options, and prognostic assessments, leading to better patient prognosis. This highlights why these methods are now considered as primary tools in rare cancer research. However, these methods still face multiple limitations, including false positive results, limited precision, and high costs.

CONCLUSION: Genetic testing and sequencing technologies have significantly advanced the field of rare cancer research by enabling the identification of key biomarkers for precision diagnosis, treatment, and prognosis. Despite existing limitations, their integration into clinical and research fields continues to improve the development of personalized medicine strategies for rare and complex cancer types.

PMID:40249565 | DOI:10.1002/cam4.70853

J Infect Dis. 2025 Apr 18:jiaf124. doi: 10.1093/infdis/jiaf124. Online ahead of print.

ABSTRACT

The study investigates therapeutic strategies for managing chronic Mycobacterium abscessus infections, particularly in people with cystic fibrosis (PWCF) who are ineligible for standard elexacaftor, tezacaftor, ivakaftor (ETI) treatments. Apoptotic body-like liposomes loaded with phosphatidylinositol 5-phosphate (ABL/PI5P) were tested in vitro in M. abscessus-infected macrophages from PWCF as potential treatment. ABL/PI5P reduced intracellular bacterial viability and showed enhanced effects on a M. abscessus clinical strain when combined with amikacin. Notably, ABL/PI5P was effective on macrophages from PWCF not receiving ETI therapy. The findings suggest ABL/PI5P liposomes as a promising alternative or adjunct therapy, especially for those who cannot access ETI treatment, warranting further clinical investigation.

PMID:40249250 | DOI:10.1093/infdis/jiaf124

Clin Transl Gastroenterol. 2025 Apr 18. doi: 10.14309/ctg.0000000000000846. Online ahead of print.

ABSTRACT

OBJECTIVE: Pathogenic variants (PVs) in the cystic fibrosis transmembrane conductance regulator (CFTR) gene are commonly reported across the spectrum of pancreatitis, including acute (AP), recurrent acute (RAP), and chronic pancreatitis (CP). We aimed to define the pooled prevalence of CFTR PVs according to pancreatitis phenotype.

METHODS: A systematic search using synonyms for "CFTR" and "pancreatitis" was performed in Embase and Pubmed databases. The primary outcome was the frequency of subjects with at least one CFTR PV amongst those who underwent germline CFTR testing. Subgroup analyses included age, pancreatitis etiology, and genetic testing strategy. Confidence intervals (CIs) were obtained using the exact binomial method (Clopper-Pearson), and a Sidik-Jonkman random effects model was used to calculate pooled prevalence.

RESULTS: In total, 138 studies were included in the final analysis; 17 (n=1,873) reported populations with AP, 21 (n=1,172) with RAP, 86 (n=13,428) with CP, and 36 (n=4,521) with unspecified pancreatitis type. The pooled prevalence of at least one CFTR PV was 8.0% (95% CI: 4.3 - 14.4%) of AP, 16.4% (95% CI: 10.2 - 25.4%) of RAP, 15.3% (95% CI: 12.2 - 19.0%) of CP, and 25.0% (95% CI: 17.5 - 34.3%) of unspecified pancreatitis. Heterogeneity was high in each phenotype (I2 value range 88.3% to 96.7%).

CONCLUSIONS: These findings underscore the complex landscape of CFTR PVs in pancreatitis, emphasizing the importance of tailored approaches in addressing this genetic component across diverse patient groups and phenotypic presentations. Additionally, these data are useful for pre-test genetic counseling, and provide a justification for developing CFTR-directed interventions.

PMID:40249094 | DOI:10.14309/ctg.0000000000000846

Abdom Radiol (NY). 2025 Apr 18. doi: 10.1007/s00261-025-04955-3. Online ahead of print.

ABSTRACT

PURPOSE: To evaluate the efficacy of deep learning reconstruction (DLR) in diffusion-weighted imaging (DWI) with single-shot echo-planar imaging (SSEPI) for endometrial cancer, compared to multiplexed sensitivity-encoding (MUSE) DWI.

METHODS: We retrospectively reviewed 31 women with surgically confirmed endometrial cancer who underwent preoperative pelvic magnetic resonance imaging (MRI) including DWI. Qualitative analysis including overall image quality, susceptibility artifacts, sharpness of the uterine edge, and lesion conspicuity were compared among conventional SSEPI (SSEPI-C), SSEPI with DLR (SSEPI-DL), and MUSE using the Friedman's test. Quantitative analysis including the apparent diffusion coefficient (ADC) values, noise, signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were also compared among three DWI sequences using the Friedman's test. In addition, the diagnostic accuracy for deep myometrial invasion was compared to three DWI sequences using Cochran's Q test.

RESULTS: The scores of overall image quality, sharpness of the uterine edge, and lesion conspicuity in SSEPI-DL were higher than SSEPI-C (p < 0.001) with no significant difference compared to MUSE (p > 0.05). Noise in SSEPI-DL was lower than SSEPI-C (p < 0.001), with no significant difference compared to MUSE (p > 0.05). SNR and CNR in SSEPI-DL were also superior to SSEPI-C (p < 0.001), and comparable to MUSE (p > 0.05). The diagnostic accuracy for detecting deep myometrial invasion showed no significant difference among SSEPI-C, SSEPI-DL and MUSE (p > 0.05).

CONCLUSION: DLR improves the image quality of DWI in endometrial cancer, demonstrating image quality equivalent to that of SSEPI-DL and MUSE. SSEPI-DL can be an alternative to MUSE in female pelvic MRI, with the benefit of significantly shortened scan time.

PMID:40249551 | DOI:10.1007/s00261-025-04955-3

Methods Mol Biol. 2025;2915:71-99. doi: 10.1007/978-1-0716-4466-9_4.

ABSTRACT

A classical problem in neuroscience, biology, and chemistry is linking the chemical structure of odorants to their olfactory perception. This difficulty arises from the subjective nature of odor perception, incomplete understanding of the physiological mechanisms involved, and the absence of standardized odor descriptions. Machine learning and other computational approaches have recently been applied to tackle this challenge. This chapter presents a comprehensive workflow for constructing machine learning models for odor prediction, covering everything from problem formulation to model evaluation and real-world deployment. We also delve into recent advancements to enhance and interpret data-driven predictions while acknowledging the current limitations. The methodology outlined here offers a valuable framework for synthetic chemists and data scientists, enabling them to address the broader issue of olfaction and cater to specific needs within the fragrance and perfume industries.

PMID:40249484 | DOI:10.1007/978-1-0716-4466-9_4

EJNMMI Res. 2025 Apr 18;15(1):42. doi: 10.1186/s13550-025-01238-2.

ABSTRACT

BACKGROUND: The quality of clinical PET/CT images is critical for both accurate diagnosis and image-based research. However, current image quality assessment (IQA) methods predominantly rely on handcrafted features and region-specific analyses, thereby limiting automation in whole-body and multicenter evaluations. This study aims to develop an expert-perceptive deep learning-based IQA system for [18F]FDG PET/CT to tackle the lack of automated, interpretable assessments of clinical whole-body PET/CT image quality.

METHODS: This retrospective multicenter study included clinical whole-body [18F]FDG PET/CT scans from 718 patients. Automated identification and localization algorithms were applied to select predefined pairs of PET and CT slices from whole-body images. Fifteen experienced experts, trained to conduct blinded slice-level subjective assessments, provided average visual scores as reference standards. Using the MANIQA framework, the developed IQA model integrates the Vision Transformer, Transposed Attention, and Scale Swin Transformer Blocks to categorize PET and CT images into five quality classes. The model's correlation, consistency, and accuracy with expert evaluations on both PET and CT test sets were statistically analysed to assess the system's IQA performance. Additionally, the model's ability to distinguish high-quality images was evaluated using receiver operating characteristic (ROC) curves.

RESULTS: The IQA model demonstrated high accuracy in predicting image quality categories and showed strong concordance with expert evaluations of PET/CT image quality. In predicting slice-level image quality across all body regions, the model achieved an average accuracy of 0.832 for PET and 0.902 for CT. The model's scores showed substantial agreement with expert assessments, achieving average Spearman coefficients (ρ) of 0.891 for PET and 0.624 for CT, while the average Intraclass Correlation Coefficient (ICC) reached 0.953 for PET and 0.92 for CT. The PET IQA model demonstrated strong discriminative performance, achieving an area under the curve (AUC) of ≥ 0.88 for both the thoracic and abdominal regions.

CONCLUSIONS: This fully automated IQA system provides a robust and comprehensive framework for the objective evaluation of clinical image quality. Furthermore, it demonstrates significant potential as an impartial, expert-level tool for standardised multicenter clinical IQA.

PMID:40249445 | DOI:10.1186/s13550-025-01238-2

J Comput Assist Tomogr. 2025 Apr 7. doi: 10.1097/RCT.0000000000001755. Online ahead of print.

ABSTRACT

OBJECTIVE: To assess image quality between single-energy CT (SECT) and dual-energy CT (DECT) scans compared with a novel deep learning (DL) reconstruction for SECT used to improve contrast enhancement.

METHODS: The raw data from a prior prospective HIPAA-compliant study (March through August 2022) was used to create a novel reconstruction in patients with biopsy-proven colorectal adenocarcinoma and liver metastases. Patients underwent 120 kVp SECT and DECT (50 keV reconstruction) abdominal scans in the portal venous phase in the same breath hold. Two readers independently assessed the scans.

RESULTS: The final study group was 13 men and 2 women with a mean age of 60 years ± 10, a mean height of 171 cm ± 8, a mean weight of 87 kg ± 23, and a mean body mass index of 30 kg/m2 ± 6. Liver, pancreas, spleen, psoas muscle, and aorta HUs were all significantly higher with the virtual DL reconstruction compared with the 120 kVp series, but significantly lower than the 50 keV series (P<0.05). Readers scored the DL reconstruction to have better contrast enhancement than the standard 120 kVp series and improved artifacts, noise texture, and resolution compared with the 50 keV series (P<0.05).

CONCLUSIONS: Contrast enhancement with the new reconstruction is superior compared with the standard 120 kVp series approaching that of 50 keV DECT, but with improved perception of artifacts, noise texture, and resolution.

PMID:40249273 | DOI:10.1097/RCT.0000000000001755

Radiol Imaging Cancer. 2025 May;7(3):e240184. doi: 10.1148/rycan.240184.

ABSTRACT

Purpose To build a deep learning framework using contrast-enhanced MRI for lesion segmentation and automatic molecular subtype classification in breast cancer. Materials and Methods This retrospective multicenter study included patients with biopsy-proven invasive breast cancer between January 2015 and January 2021. An automatic breast lesion segmentation model was developed using three-dimensional (3D) ResU-Net as the backbone, and its accuracy was evaluated in an internal and two external testing datasets using the Dice score. An ensemble model for classification of breast cancer into four molecular subtypes (Ensemble ResNet) was then developed by combining both two-dimensional and 3D lesion features. The performance of Ensemble ResNet was evaluated in the three testing datasets using the area under the receiver operating characteristic curve (AUC). Results A total of 687 female patients (mean age ± SD, 48.70 years ± 8.97) were included, with 289, 61, 73, and 264 patients included in the training, internal testing, and two external testing datasets, respectively. The proposed segmentation model achieved high accuracy in internal testing dataset 1, external testing dataset 2, and external testing dataset 3 (Dice scores: 0.86, 0.82, 0.85) and luminal A, luminal B, human epidermal growth factor receptor 2 (HER2)-enriched, and triple-negative breast cancer (TNBC) subtypes (Dice scores: 0.8571, 0.8323, 0.8199, 0.8481). Ensemble ResNet demonstrated high performance for the prediction of luminal A subtypes (AUC range, 0.74-0.84), luminal B subtypes (AUC range, 0.68-0.72), HER2-enriched subtypes (AUC range, 0.73-0.82), and TNBC (AUC range, 0.80-0.81) in the three testing datasets. Conclusion The proposed novel deep learning framework based on MRI achieved high, robust performance in fully automatic classification of breast cancer molecular subtypes. Keywords: MR-Imaging, Breast, Oncology, Breast Cancer, Molecular Subtype, Deep Learning Framework Supplemental material is available for this article. © RSNA, 2025.

PMID:40249269 | DOI:10.1148/rycan.240184

J Natl Cancer Inst. 2025 Apr 18:djaf103. doi: 10.1093/jnci/djaf103. Online ahead of print.

ABSTRACT

BACKGROUND: European studies suggest artificial intelligence (AI) can reduce interval breast cancers (IBCs). However, research on IBC classification and AI's effectiveness in the U.S., particularly using digital breast tomosynthesis (DBT) and annual screening, is limited. We aimed to mammographically classify IBCs and assess AI performance using a 12-month screening interval.

METHODS: From digital mammography (DM) and DBT screening mammograms acquired 2010-2019 at a U.S. tertiary care academic center, we identified IBCs diagnosed <12 months after a negative mammogram. At least three breast radiologists retrospectively classified IBCs as missed-reading error, minimal signs-actionable, minimal signs-non-actionable, true interval, occult, or missed-technical error. A deep-learning AI tool assigned risk scores (1-10) to the negative index screening mammograms, with scores ≥8 considered "flagged." Statistical analysis evaluated associations among IBC types and AI exam scores, AI markings, and patient/tumor characteristics.

RESULTS: From 184,935 screening mammograms (65% DM, 35% DBT), we identified 148 IBCs in 148 women (mean age, 61±12 years). Of these, 26% were minimal signs-actionable; 24% occult; 22% minimal signs-non-actionable; 17% missed-reading error; 6% true interval; and 5% missed-technical error (p<.001). AI scored 131 mammograms (17 errors excluded). AI most frequently flagged exams with missed-reading errors (90%), minimal signs-actionable (89%) and minimal signs-non-actionable (72%) (p=.02). AI localized mammographically-visible types more accurately (35-68%) than non-visible types (0-50%, p=.02).

CONCLUSION: AI more frequently flagged and accurately localized IBC types that were mammographically visible at screening (missed or minimal signs), as compared to true interval or occult cancers.

PMID:40249223 | DOI:10.1093/jnci/djaf103

Top Magn Reson Imaging. 2025 Apr 21;34(1):e0316. doi: 10.1097/RMR.0000000000000316. eCollection 2025 Jun 1.

ABSTRACT

OBJECTIVES: To develop and evaluate a deep learning technique for the differentiation of hepatocellular carcinoma (HCC) using "simplified intravoxel incoherent motion (IVIM) parameters" derived from only 3 b-value images.

MATERIALS AND METHODS: Ninety-eight retrospective magnetic resonance imaging data were collected (68 men, 30 women; mean age 59 ± 14 years), including T2-weighted imaging with fat suppression, in-phase, out-of-phase, and diffusion-weighted imaging (b = 0, 100, 800 s/mm2). Ninety percent of data were used for stratified 10-fold cross-validation. After data preprocessing, diffusion-weighted imaging images were used to compute simplified IVIM and apparent diffusion coefficient (ADC) maps. A 17-layer 3D convolutional neural network (3D-CNN) was implemented, and the input channels were modified for different strategies of input images.

RESULTS: The 3D-CNN with IVIM maps (ADC, f, and D*) demonstrated superior performance compared with other strategies, achieving an accuracy of 83.25 ± 6.24% and area under the receiver-operating characteristic curve of 92.70 ± 8.24%, significantly surpassing the baseline of 50% (P < 0.05) and outperforming other strategies in all evaluation metrics. This success underscores the effectiveness of simplified IVIM parameters in combination with a 3D-CNN architecture for enhancing HCC differentiation accuracy.

CONCLUSIONS: Simplified IVIM parameters derived from 3 b-values, when integrated with a 3D-CNN architecture, offer a robust framework for HCC differentiation.

PMID:40249154 | DOI:10.1097/RMR.0000000000000316

J Forensic Sci. 2025 Apr 18. doi: 10.1111/1556-4029.70058. Online ahead of print.

ABSTRACT

The detection of cannabis and cannabis-related products is a critical task for forensic laboratories and law enforcement agencies, given their harmful effects. Forensic laboratories analyze large quantities of plant material annually to identify genuine cannabis and its illicit substitutes. Ensuring accurate identification is essential for supporting judicial proceedings and combating drug-related crimes. The naked eye alone cannot distinguish between genuine cannabis and non-cannabis plant material that has been sprayed with synthetic cannabinoids, especially after distribution into the market. Reliable forensic identification typically requires two colorimetric tests (Duquenois-Levine and Fast Blue BB), as well as a drug laboratory expert test for affirmation or negation of cannabis hair (non-glandular trichomes), making the process time-consuming and resource-intensive. Here, we propose a novel deep learning-based computer vision method for identifying non-glandular trichome hairs in cannabis. A dataset of several thousand annotated microscope images was collected, including genuine cannabis and non-cannabis plant material apparently sprayed with synthetic cannabinoids. Ground-truth labels were established using three forensic tests, two chemical assays, and expert microscopic analysis, ensuring reliable classification. The proposed method demonstrated an accuracy exceeding 97% in distinguishing cannabis from non-cannabis plant material. These results suggest that deep learning can reliably identify non-glandular trichome hairs in cannabis based on microscopic trichome features, potentially reducing reliance on costly and time-consuming expert microscopic analysis. This framework provides forensic departments and law enforcement agencies with an efficient and accurate tool for identifying non-glandular trichome hairs in cannabis, supporting efforts to combat illicit drug trafficking.

PMID:40249026 | DOI:10.1111/1556-4029.70058

Plant J. 2025 Apr;122(1):e70160. doi: 10.1111/tpj.70160.

ABSTRACT

Willows can alleviate soil salinisation while generating sustainable feedstock for biorefinery, yet the metabolomic adaptations underlying their tolerance remain poorly understood. Salix miyabeana was treated with two environmentally abundant salts, NaCl and Na2SO4, in a 12-week pot trial. Willows tolerated salts across all treatments (up to 9.1 dS m-1 soil ECe), maintaining biomass while selectively partitioning ions, confining Na+ to roots and accumulating Cl- and SO 4 2 - $$ {\mathrm{SO}}_4^{2-} $$ in the canopy and adapting to osmotic stress via reduced stomatal conductance. Untargeted metabolomics captured >5000 putative compounds, including 278 core willow metabolome compounds constitutively produced across organs. Across all treatments, salinity drove widespread metabolic reprogramming, altering 28% of the overall metabolome, with organ-tailored strategies. Comparing salt forms at equimolar sodium, shared differentially abundant metabolites were limited to 3% of the metabolome, representing the generalised salinity response, predominantly in roots. Anion-specific metabolomic responses were extensive. NaCl reduced carbohydrates and tricarboxylic acid cycle intermediates, suggesting potential carbon and energy resource pressure, and accumulated root structuring compounds, antioxidant flavonoids, and fatty acids. Na2SO4 salinity triggered accumulation of sulphur-containing larger peptides, suggesting excess sulphate incorporation leverages ion toxicity to produce specialised salt-tolerance-associated metabolites. This high-depth picture of the willow metabolome underscores the importance of capturing plant adaptations to salt stress at organ scale and considering ion-specific contributions to soil salinity.

PMID:40249060 | DOI:10.1111/tpj.70160

J Chem Inf Model. 2025 Apr 18. doi: 10.1021/acs.jcim.5c00165. Online ahead of print.

ABSTRACT

Here, we introduce the Structural Systems Biology (SSB) toolkit, a Python library that integrates structural macromolecular data with systems biology simulations to model signal-transduction pathways of G-protein-coupled receptors (GPCRs). Our framework streamlines simulation and analysis of the mathematical models of GPCRs cellular pathways, facilitating the exploration of the signal-transduction kinetics induced by ligand-GPCR interactions: the dose-response of the ligand can be modeled, along with the corresponding change in the concentration of other signaling molecular species over time, like for instance [Ca2+] or [cAMP]. SSB toolkit brings to light the possibility of easily investigating the subcellular effects of ligand binding on receptor activation, even in the presence of genetic mutations, thereby enhancing our understanding of the intricate relationship between ligand-target interactions at the molecular level and the higher-level cellular and (patho)physiological response mechanisms.

PMID:40248991 | DOI:10.1021/acs.jcim.5c00165