BMC Pulm Med. 2025 Apr 5;25(1):157. doi: 10.1186/s12890-025-03624-x.

ABSTRACT

BACKGROUND: In this research project, we examined the relationship between tracheal size and respiratory function in individuals with Idiopathic Pulmonary Fibrosis (IPF). IPF is a long-term condition that affects the functioning of the lungs.

METHODS: This retrospective study included 86 patients diagnosed with IPF. Tracheal and bronchial diameters were measured using high-resolution computed tomography (HRCT) and pulmonary function tests (PFTs); Force vital capacity (FVC), diffusion capacity for carbon monoxide (DLCO) and the gender, age, physiology (GAP) index was calculated. Patients were grouped according to demographic characteristics such as age, gender and smoking.

RESULTS: There was a significant positive correlation between the anteroposterior (AP) and transverse diameters of the trachea in the subcricoid region and the GAP index (r = 0.318, p = 0.003 and r = 0.312, p = 0.004, respectively). Similarly, subcricoid and carina areas were significantly correlated with both GAP index (r = 0.307, p = 0.006 and r = 0.334, p = 0.003, respectively) and FVC/DLCO ratio (r = 0.218, p = 0.049 and r = 0.245, p = 0.027, respectively). The main bronchial areas were also positively correlated with the GAP index, but no significant correlation was found between FVC and DLCO values and airway measurements. Each unit increase in GAP index was associated with a 1.69-fold increase in mortality risk (p = 0.0016, 95% confidence interval: 1.22-2.34).

CONCLUSION: Tracheal and main bronchial areas can be used as potential biomarkers in the assessment of disease severity and prognosis in IPF patients. In particular, the significant correlation of subcricoid and carina areas with both GAP index and FVC/DLCO ratio suggests that these measurements may be useful in the evaluation of disease progression.

PMID:40188355 | DOI:10.1186/s12890-025-03624-x

Sci Rep. 2025 Apr 5;15(1):11730. doi: 10.1038/s41598-025-94575-0.

ABSTRACT

There are well-documented differences in idiopathic pulmonary fibrosis (IPF) between sexes. The sex-specific prevalence of interstitial lung disease (ILD) subtypes in patients who require a full diagnostic work-up, including transbronchial cryobiopsy (TCB), after initial multidisciplinary discussion (MDD) is still unknown. Retrospective analysis of sex dispareties in patients with ILD who received an interdisciplinary indication for lung biopsy and underwent bronchoalveolar lavage, TCB and, if necessary, surgical lung biopsy at our ILD centre in Heidelberg between 11/17 and 12/21. The analysis included clinical parameters, visual assessment of computed tomography (CT), automated histogram analyses of lung density by validated software and final MDD-ILD classifications. A total of 402 patients (248 men, 154 women; mean age 68 ± 12 years) were analysed. Smoking behaviour was similar between the sexes, but women were more exposed to environmental factors, whereas men were more exposed to occupational factors. Women had higher rates of thyroid disease (29.9% vs. 12.5%; p < 0.001) and extrathoracic malignancies (16.2% vs. 9.3%; p = 0.041), but lower rates of coronary heart disease (7.1% vs. 19.8%; p < 0.001), stroke (1.3% vs. 6.5%; p = 0.014) and sleep apnoea (5.8% vs. 17.7%; p < 0.001). There were no sex differences regarding CT lung density. On visual inspection, women were less likely to have reticular opacities (65% vs. 76%; p = 0.017) and features of usual interstitial pneumonia (17% vs. 34%; p < 0.001). Among final diagnoses, hypersensitivity pneumonitis was more common in women (34.4%) compared to men (21.8%; p = 0.007). In contrast, IPF was more common in men (22.6%) than in women (7.1%; p < 0.001), and unclassifiable interstitial lung disease was also more frequent in men (21.8%) compared to women (6.5%; p < 0.001). This study highlights significant sex-based differences in the prevalence and characteristics of ILD requiring comprehensive diagnostic work-up. These findings underscore the importance of considering sex-specific factors in the diagnosis and management of ILD.

PMID:40188253 | DOI:10.1038/s41598-025-94575-0

Gene. 2025 Apr 3:149464. doi: 10.1016/j.gene.2025.149464. Online ahead of print.

ABSTRACT

Idiopathic pulmonary fibrosis (IPF) is a progressive lung disorder that is characterized by the disruption of lung architecture and respiratory failure. Notwithstanding the advent of novel therapeutic agents such as pirfenidone and nintedanib, there remains a pressing need for the development of innovative diagnostic and therapeutic strategies. Next-generation sequencing allows for the analysis of gene expression and the discovery of biomarkers. The objective of our study was to identify IPF-specific gene signatures, construct a diagnostic nomogram, and explore the role of the extracellular matrix (ECM) and epithelial-to-mesenchymal transition (EMT) in IPF pathogenesis. Utilizing data from the Gene Expression Omnibus (GEO) database, we identified differentially expressed genes (DEGs), performed weighted correlation network analysis (WGCNA), and constructed a nomogram. The present study has identified a group of key genes that are associated with IPF. The identified genes include GREM1, ITLN2, MAP3K15, RGS9BP, and SLCO1A2. The results of the immunohistochemical analysis indicated a significant correlation between these central genes and immune cell infiltration. Furthermore, Gene Set Enrichment Analysis (GSEA) revealed that these genes play a critical role in the pathogenesis of IPF. To validate the diagnostic potential of these core genes, we performed confirmatory analyses in independent Gene Expression Omnibus (GEO) datasets. We observed a significant upregulation of GREM1 expression in IPF animal and cellular models. These findings provide new insights into the molecular mechanisms of IPF and suggest potential targets for future diagnostic and therapeutic strategies.

PMID:40187620 | DOI:10.1016/j.gene.2025.149464

Sci Data. 2025 Apr 5;12(1):579. doi: 10.1038/s41597-025-04854-8.

ABSTRACT

The spotted lanternfly (Lycorma delicatula) is a planthopper species (Hemiptera: Fulgoridae) native to China but invasive in South Korea, Japan, and the United States where it is a significant threat to agriculture. Genomic resources are critical to both management of this species and understanding the genomic characteristics of successful invaders. We report an annotated, haplotype-phased, chromosome-level genome assembly for the spotted lanternfly using PacBio long-read sequencing, Hi-C technology, and RNA-seq. The 2.2 Gbp genome comprises 13 chromosomes, and whole genome resequencing of eighty-two adults indicated chromosome four as the sex chromosome and a corresponding XO sex-determination system. We identified over 12,000 protein-coding genes and performed functional annotation, facilitating the identification of candidate genes that may hold importance for spotted lanternfly control. The assemblies and annotations were highly complete with over 96% of BUSCO genes complete regardless of the database (i.e., Eukaryota, Arthropoda, Insecta). This reference-quality genome will serve as an important resource for development and optimization of management practices for the spotted lanternfly and invasive species genomics as a whole.

PMID:40188159 | DOI:10.1038/s41597-025-04854-8

Nat Commun. 2025 Apr 5;16(1):3268. doi: 10.1038/s41467-025-58615-7.

ABSTRACT

Ubiquitylation of H2A (H2AK119ub1) by the polycomb repressive complexe-1 plays a key role in the initiation of facultative heterochromatin formation in somatic cells. Here we evaluate the contribution of sperm derived H2AK119ub1 to embryo development. In Xenopus laevis we found that H2AK119ub1 is present during spermiogenesis and into early embryonic development, highlighting its credential for a role in the transmission of epigenetic information from the sperm to the embryo. In vitro treatment of sperm with USP21, a H2AK119ub1 deubiquitylase, just prior to injection to egg, results in developmental defects associated with gene upregulation. Sperm H2AK119ub1 editing disrupts egg factor mediated paternal chromatin remodelling processes. It leads to post-replication accumulation of H2AK119ub1 on repeat element of the genome instead of CpG islands. This shift in post-replication H2AK119ub1 distribution triggered by sperm epigenome editing entails a loss of H2AK119ub1 from genes misregulated in embryos derived from USP21 treated sperm. We conclude that sperm derived H2AK119ub1 instructs egg factor mediated epigenetic remodelling of paternal chromatin and is required for embryonic development.

PMID:40188103 | DOI:10.1038/s41467-025-58615-7

J Mol Biol. 2025 Apr 3:169124. doi: 10.1016/j.jmb.2025.169124. Online ahead of print.

ABSTRACT

The RNA World hypothesis predicts that self-replicating RNAs evolved before DNA genomes and coded proteins. Despite widespread support for the RNA World, self-replicating RNAs have yet to be identified in a natural context, leaving a key 'missing link' for this explanation of the origin of life. Inspired by recent work showing that condensates of charged polymers are capable of catalyzing chemical reactions, we consider a catalytic RNA condensate as a candidate for the self-replicating RNA. Specifically, we propose that short, low-complexity RNA polymers formed catalytic condensates capable of templated RNA polymerization. Because the condensate properties depend on the RNA sequences, RNAs that formed condensates with improved polymerization and demixing capacity would be amplified, leading to a 'condensate chain reaction' and evolution by natural selection. Many of the needed properties of this self-replicating RNA condensate have been realized experimentally in recent studies and our predictions could be tested with current experimental and theoretical tools. Our theory addresses central problems in the origins of life: (i) the origin of compartmentalization, (ii) the error threshold for the accuracy of templated replication, (iii) the free energy cost of maintaining an information-rich population of replicating RNA polymers. Furthermore, we note that the extant nucleolus appears to satisfy many of the requirements of an evolutionary relic for the model we propose. More generally, we suggest that future work on the origin of life would benefit from condensate-centric biophysical models of RNA evolution.

PMID:40187684 | DOI:10.1016/j.jmb.2025.169124

Cell. 2025 Apr 2:S0092-8674(25)00283-1. doi: 10.1016/j.cell.2025.03.010. Online ahead of print.

ABSTRACT

Despite the fundamental importance of gut microbes, the genetic basis of their colonization remains largely unexplored. Here, by applying cross-species genotype-habitat association at the tree-of-life scale, we identify conserved microbial gene modules associated with gut colonization. Across thousands of species, we discovered 79 taxonomically diverse putative colonization factors organized into operonic and non-operonic modules. They include previously characterized colonization pathways such as autoinducer-2 biosynthesis and novel processes including tRNA modification and translation. In vivo functional validation revealed YigZ (IMPACT family) and tRNA hydroxylation protein-P (TrhP) are required for E. coli intestinal colonization. Overexpressing YigZ alone is sufficient to enhance colonization of the poorly colonizing MG1655 E. coli by >100-fold. Moreover, natural allelic variations in YigZ impact inter-strain colonization efficiency. Our findings highlight the power of large-scale comparative genomics in revealing the genetic basis of microbial adaptations. These broadly conserved colonization factors may prove critical for understanding gastrointestinal (GI) dysbiosis and developing therapeutics.

PMID:40187346 | DOI:10.1016/j.cell.2025.03.010

Pharm Dev Technol. 2025 Apr 5:1-33. doi: 10.1080/10837450.2025.2487575. Online ahead of print.

ABSTRACT

The COVID-19 pandemic exposed the fragility of today's marketed treatments for respiratory infections. As a primary site of infection, the upper airways may represent a key access route for the control and treatment for these conditions. The present study aims to explore and identify, through a patent review, the novelty of therapies for COVID-19 that use the intranasal route for drug administration. A search was carried out in Wipo and Espacenet, using the descriptors "COVID-19 OR SARS-CoV 2" AND "treatment OR therapy" AND NOT "vaccine OR immunizing" and the classification "A61K9/0043". Of the 151 patents identified, we excluded 73 duplicates, and 36 documents that meet the criteria adopted for exclusion (not nasally administered formulations, vaccines, post COVID-19 treatments, uncertain route of administration or form). We identified 78 unique patents on patent databases, of which 42 were selected for this review. The documents revealed the use of the intranasal pathway not only for drug repositioning but also for using plant-derived and biological molecules. Overall, the new formulations explore a variety of known drugs and natural products incorporated in drug carrier systems and devices for drug delivery and administration. Thus, the intranasal route remains a promising strategy for drug delivery, offering direct access to the primary infection site and warranting further exploration.

PMID:40186505 | DOI:10.1080/10837450.2025.2487575

Eur J Pediatr. 2025 Apr 5;184(5):281. doi: 10.1007/s00431-025-06101-z.

ABSTRACT

The diagnosis and treatment of rare diseases present significant global challenges. This study aimed to identify the difficulties faced by specialists in the diagnosis and management of rare diseases, as well as to gather their recommendations for potential solutions. An expert committee specializing in inborn metabolic disease and genetics developed a comprehensive survey, which was then distributed online to professionals working with rare diseases. A total of 21 specialists actively engaged in the management of rare diseases participated in the survey. All participants acknowledged the substanstial significant diagnostic challenges associated with rare diseases, with 86% indicating that these diagnostic challenges negatively affect their clinical practice. The primary obstacles encountered in the diagnosis and follow-up of rare diseases were low awareness, a lack of a multidisciplinary approach, insufficient numbers of specialists and inadequate infrastructure, limited newborn screening programs, challenges in accessing treatment, and insufficient psychosocial support. All participants emphasized the need for a multidisciplinary approach in the management of rare diseases. Proposed solutions included enhanced training for healthcare professionals, the establishment of multidisciplinary teams and diagnostic algorithms, the regular convening of councils and meetings, and the establishment of robust registries. While all participants rated their own clinical experience as proficient in diagnosing and treating rare diseases, the establishment of multidisciplinary teams was the most frequently suggested area for improvement.

CONCLUSION: Addressing the challenges in the diagnosis, treatment, and monitoring of rare diseases requires a multifaceted approach, including raising awareness, enhancing patient services, developing robust research and improving the infrastructure, establishing multidisciplinary care frameworks, and implementing preventive medicine and social policies.

WHAT IS KNOWN: • It is estimated that over 300 million people globally are living with one or more rare diseases. The process of diagnosis, treatment, and follow-up of rare diseases involves significant global challenges.

WHAT IS NEW: • In our study, the difficulties encountered by specialists in the diagnosis and treatment of rare diseases in Türkiye and solution suggestions are presented. This is the first study on this subject in Türkiye.

PMID:40186762 | DOI:10.1007/s00431-025-06101-z

Int J Comput Assist Radiol Surg. 2025 Apr 5. doi: 10.1007/s11548-025-03348-7. Online ahead of print.

ABSTRACT

PURPOSE: Metal artifacts remain a persistent issue in intraoperative CBCT imaging. Particularly in orthopedic and trauma applications, these artifacts obstruct clinically relevant areas around the implant, reducing the modality's clinical value. Metal artifact avoidance (MAA) methods have shown potential to improve image quality through trajectory adjustments, but often fail in clinical practice due to their focus on irrelevant objects and high computational demands. To address these limitations, we introduce the novel parametric metal artifact avoidance (P-MAA) method.

METHODS: The P-MAA method first detects keypoints in two scout views using a deep learning model. These keypoints are used to model each clinically relevant object as an ellipsoid, capturing its position, extent, and orientation. We hypothesize that fine details of object shapes are less critical for artifact reduction. Based on these ellipsoidal representations, we devise a computationally efficient metric for scoring view trajectories, enabling fast, CPU-based optimization. A detection model for object localization was trained using both simulated and real data and validated on real clinical cases. The scoring method was benchmarked against a raytracing-based approach.

RESULTS: The trained detection model achieved a mean average recall of 0.78, demonstrating generalizability to unseen clinical cases. The ellipsoid-based scoring method closely approximated results using raytracing and was effective in complex clinical scenarios. Additionally, the ellipsoid method provided a 33-fold increase in speed, without the need for GPU acceleration.

CONCLUSION: The P-MAA approach provides a feasible solution for metal artifact avoidance in CBCT imaging, enabling fast trajectory optimization while focusing on clinically relevant objects. This method represents a significant step toward practical intraoperative implementation of MAA techniques.

PMID:40186717 | DOI:10.1007/s11548-025-03348-7

Pediatr Radiol. 2025 Apr 5. doi: 10.1007/s00247-025-06225-2. Online ahead of print.

ABSTRACT

BACKGROUND: Hepatocyte cytokeratin 7 (CK7) is a reliable marker for evaluating the severity of cholestasis in chronic cholestatic cholangiopathies. However, there is currently no noninvasive test available to assess the status of hepatocyte CK7 in pancreaticobiliary maljunction patients.

OBJECTIVE: We aimed to develop a deep learning radiomics nomogram using magnetic resonance images (MRIs) to preoperatively identify the hepatocyte CK7 status and assess cholestasis progression in patients with pancreaticobiliary maljunction.

MATERIALS AND METHODS: In total, 180 pancreaticobiliary maljunction patients were retrospectively enrolled and were randomly divided into a training cohort (n = 144) and a validation cohort (n = 36). CK7 status was determined through immunohistochemical analysis. Pyradiomics and pretrained ResNet50 were used to extract radiomics and deep learning features, respectively. To construct the radiomics and deep learning signature, feature selection methods including the minimum redundancy-maximum relevance and least absolute shrinkage and selection operator were employed. The integrated deep learning radiomics nomogram model was constructed by combining the imaging signatures and valuable clinical feature.

RESULTS: The deep learning signature exhibited superior predictive performance compared with the radiomics signature, as evidenced by the higher area under the curve (AUC) values in validation cohort (0.92 vs. 0.81). Further, the deep learning radiomics nomogram, which incorporated the radiomics signature, deep learning signature, and Komi classification, demonstrated excellent predictive ability for CK7 expression, with AUC value of 0.95 in the validation cohort.

CONCLUSION: The proposed deep learning radiomics nomogram exhibits promising performance in accurately identifying hepatic CK7 expression, thus facilitating prediction of cholestasis progression and perhaps earlier initiation of treatment in pancreaticobiliary maljunction children.

PMID:40186654 | DOI:10.1007/s00247-025-06225-2

Pediatr Radiol. 2025 Apr 5. doi: 10.1007/s00247-025-06230-5. Online ahead of print.

ABSTRACT

BACKGROUND: Radial k-space sampling is widely employed in paediatric magnetic resonance imaging (MRI) to mitigate motion and aliasing artefacts. Artificial intelligence (AI)-based image reconstruction has been developed to enhance image quality and accelerate acquisition time.

OBJECTIVE: To assess image quality of deep learning (DL)-based denoising image reconstruction of body MRI in children.

MATERIALS AND METHODS: Children who underwent thoraco-abdominal MRI employing radial k-space filling technique (PROPELLER) with conventional and DL-based image reconstruction between April 2022 and January 2023 were eligible for this retrospective study. Only cases with previous MRI including comparable PROPELLER sequences with conventional image reconstruction were selected. Image quality was compared between DL-reconstructed axial T1-weighted and T2-weighted images and conventionally reconstructed images from the same PROPELLER acquisition. Quantitative image quality was assessed by signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of the liver and spleen. Qualitative image quality was evaluated by three observers using a 4-point Likert scale and included presence of noise, motion artefact, depiction of peripheral lung vessels and subsegmental bronchi at the lung bases, sharpness of abdominal organ borders, and visibility of liver and spleen vessels. Image quality was compared with the Wilcoxon signed-rank test. Scan time length was compared to prior MRI obtained with conventional image reconstruction.

RESULTS: In 21 children (median age 7 years, range 1.5 years to 15.8 years) included, the SNR and CNR of the liver and spleen on T1-weighted and T2-weighted images were significantly higher with DL-reconstruction (P<0.001) than with conventional reconstruction. The DL-reconstructed images showed higher overall image quality, with improved delineation of the peripheral vessels and the subsegmental bronchi in the lung bases, sharper abdominal organ margins and increased visibility of the peripheral vessels in the liver and spleen. Not respiratory-gated DL-reconstructed T1-weighted images demonstrated more pronounced respiratory motion artefacts in comparison to conventional reconstruction (P=0.015), while there was no difference for the respiratory-gated T2-weighted images. The median scan time per slice was reduced from 6.3 s (interquartile range, 4.2 - 7.0 s) to 4.8 s (interquartile range, 4.4 - 4.9 s) for the T1-weighted images and from 5.6 s (interquartile range, 5.4 - 5.9 s) to 4.2 s (interquartile range, 3.9 - 4.8 s) for the T2-weighted images.

CONCLUSION: DL-based denoising image reconstruction of paediatric body MRI sequences employing radial k-space sampling allowed for improved overall image quality at shorter scan times. Respiratory motion artefacts were more pronounced on ungated T1-weighted images.

PMID:40186652 | DOI:10.1007/s00247-025-06230-5

Graefes Arch Clin Exp Ophthalmol. 2025 Apr 5. doi: 10.1007/s00417-025-06804-x. Online ahead of print.

ABSTRACT

PURPOSE: Given the significance and potential risks associated with Ocular Surface Squamous Neoplasia (OSSN) and the importance of its differentiation from other conditions, we aimed to develop a Deep Learning (DL) model differentiating OSSN from pterygium (PTG) using slit photographs.

METHODS: A dataset comprising slit photographs of 162 patients including 77 images of OSSN and 85 images of PTG was assembled. After manual segmentation of the images, a Python-based transfer learning approach utilizing the EfficientNet B7 network was employed for automated image segmentation. GoogleNet, a pre-trained neural network was used to categorize the images into OSSN or PTG. To evaluate the performance of our DL model, K-Fold 10 Cross Validation was implemented, and various performance metrics were measured.

RESULTS: There was a statistically significant difference in mean age between the OSSN (63.23 ± 13.74 years) and PTG groups (47.18 ± 11.53) (P-value =.000). Furthermore, 84.41% of patients in the OSSN group and 80.00% of the patients in the PTG group were male. Our classification model, trained on automatically segmented images, demonstrated reliable performance measures in distinguishing OSSN from PTG, with an Area Under Curve (AUC) of 98%, sensitivity, F1 score, and accuracy of 94%, and a Matthews Correlation Coefficient (MCC) of 88%.

CONCLUSIONS: This study presents a novel DL model that effectively segments and classifies OSSN from PTG images with a relatively high accuracy. In addition to its clinical use, this model can be potentially used as a telemedicine application.

PMID:40186633 | DOI:10.1007/s00417-025-06804-x

Am J Orthod Dentofacial Orthop. 2025 Apr 5:S0889-5406(25)00101-5. doi: 10.1016/j.ajodo.2025.02.014. Online ahead of print.

ABSTRACT

INTRODUCTION: Machine learning, a common artificial intelligence technology in medical image analysis, enables computers to learn statistical patterns from pairs of data and annotated labels. Supervised learning in machine learning allows the computer to predict how a specific anatomic structure should be segmented in new patients. This study aimed to develop and validate a deep learning algorithm that automatically creates 3-dimensional surface models of human teeth from a cone-beam computed tomography scan.

METHODS: A multiresolution dataset, including 216 × 272 × 272, 512 × 512 × 512, and 576 × 768 × 768. Ground truth labels for teeth segmentation were generated. Random partitioning was applied to allocate 140 patients to the training set, 40 to the validation set, and 30 scans for testing and model performance evaluation. Different evaluation metrics were used for assessment.

RESULTS: Our teeth identification model has achieved an accuracy of 87.92% ± 4.43% on the test set. The general (binary) teeth segmentation model achieved a notably higher accuracy, segmenting the teeth with 93.16% ± 1.18%.

CONCLUSIONS: The success of our model not only validates the efficacy of using artificial intelligence for dental imaging analysis but also sets a promising foundation for future advancements in automated and precise dental segmentation techniques.

PMID:40186597 | DOI:10.1016/j.ajodo.2025.02.014

J Appl Clin Med Phys. 2025 Apr 5:e70089. doi: 10.1002/acm2.70089. Online ahead of print.

ABSTRACT

BACKGROUND: The use of deep learning-based auto-contouring algorithms in various treatment planning services is increasingly common. There is a notable deficit of commercially or publicly available models trained on large or diverse datasets containing high-dose-rate (HDR) brachytherapy treatment scans, leading to poor performance on images that include HDR implants.

PURPOSE: To implement and evaluate automatic organs-at-risk (OARs) segmentation models for use in prostatic-and-gynecological computed tomography (CT)-guided high-dose-rate brachytherapy treatment planning.

METHODS AND MATERIALS: 1316 computed tomography (CT) scans and corresponding segmentation files from 1105 prostatic-or-gynecological HDR patients treated at our institution from 2017 to 2024 were used for model training. Data sources comprised six CT scanners including a mobile CT unit with previously reported susceptibility to image streaking artifacts. Two UNet-derived model architectures, UNet++ and nnU-Net, were investigated for bladder and rectum model training. The models were tested on 100 CT scans and clinically-used segmentation files from 62 prostatic-or-gynecological HDR brachytherapy patients, disjoint from the training set, collected in 2024. Performance was evaluated using the Dice-Similarity-Coefficient (DSC) between model predicted contours and clinically-used contours on slices in common with the Clinical-Target-Volume (CTV). Additionally, a blinded evaluation of ten random test cases was conducted by three experienced planners.

RESULTS: Median (interquartile range) 3D DSC on CTV-containing slices were 0.95 (0.04) and 0.87 (0.09) for the UNet++ bladder and rectum models, respectively, and 0.96 (0.03) and 0.88 (0.10) for the nnU-Net. The rank-sum test did not reveal statistically significant differences in these DSC (p = 0.15 and 0.27, respectively). The blinded evaluation scored trained models higher than clinically-used contours.

CONCLUSION: Both UNet-derived architectures perform similarly on the bladder and rectum and are adequately accurate to reduce contouring time in a review-and-edit context during HDR brachytherapy planning. The UNet++ models were chosen for implementation at our institution due to lower computing hardware requirements and are in routine clinical use.

PMID:40186596 | DOI:10.1002/acm2.70089

Int J Immunopathol Pharmacol. 2025 Jan-Dec;39:3946320251331791. doi: 10.1177/03946320251331791. Epub 2025 Apr 5.

ABSTRACT

Currently, no glucocorticoid dose prediction model is available for clinical practice. This study aimed to utilise machine learning techniques to develop and validate personalised dosage models. Participants were patients with SLE who were registered at Nanfang Hospital and received prednisone. Univariate analysis was used to confirm the feature variables. Subsequently, the random forest (RF) algorithm was utilised to interpolate the absent values of the feature variables. Finally, we assessed the prediction capabilities of 11 machine learning and deep-learning algorithms (Logistic, SVM, RF, Adaboost, Bagging, XGBoost, LightGBM, CatBoost, MLP, and TabNet). Finally, a confusion matrix was used to validate the three regimens. In total, 129 patients met the inclusion criteria. The XGBoost algorithm was selected as the preferred method because of its superior performance, achieving an accuracy of 0.81. The factors exhibiting the highest correlation with the prednisone dose were CYP3A4 (rs4646437), albumin (ALB), haemoglobin (HGB), anti-double-stranded DNA antibodies (Anti-dsDNA), erythrocyte sedimentation rate (ESR), age, and HLA-DQA1 (rs2187668). Based on validation, the precision and recall rates for low-dose prednisone (⩾5 mg but <7.5 mg/d) were 100% and 40% respectively. Similarly, for medium-dose prednisone (⩾7.5 mg but <30 mg/d), the accuracy and recall rates were 88% and 88%, and for high-dose prednisone (⩾30 mg but ⩽100 mg/d), the accuracy and recall rates were 62% and 100% respectively. A robust machine learning model was developed to accurately predict prednisone dosage by integrating the identified genetic and clinical factors.

PMID:40186486 | DOI:10.1177/03946320251331791

Acta Ophthalmol. 2025 Apr 4. doi: 10.1111/aos.17495. Online ahead of print.

ABSTRACT

PURPOSE: To validate a deep learning (DL) framework for detecting and quantifying cystoid fluid collections (CFC) on spectral-domain optical coherence tomography (SD-OCT) in X-linked retinoschisis (XLRS) patients.

METHODS: A no-new-U-Net model was trained using 112 OCT volumes from the RETOUCH challenge (70 for training and 42 for internal testing). External validation involved 37 SD-OCT scans from 20 XLRS patients, including 20 randomly sampled B-scans and 17 manually selected central B-scans. Three graders manually delineated the CFC on these B-scans in this external test set. The model's efficacy was evaluated using Dice and intraclass correlation coefficient (ICC) scores, assessed exclusively on the test set comprising B-scans from XLRS patients.

RESULTS: For the randomly sampled B-scans, the model achieved a mean Dice score of 0.886 (±0.010), compared to 0.912 (±0.014) for the observers. For the manually selected central B-scans, the Dice scores were 0.936 (±0.012) for the model and 0.946 (±0.012) for the graders. ICC scores between the model and reference were 0.945 (±0.014) for the randomly selected and 0.964 (±0.011) for the manually selected B-scans. Among the graders, ICC scores were 0.979 (±0.008) and 0.981 (±0.011), respectively.

CONCLUSIONS: Our validated DL model accurately segments and quantifies CFC on SD-OCT in XLRS, paving the way for reliable monitoring of structural changes. However, systematic overestimation by the DL model was observed, highlighting a key limitation for future refinement.

PMID:40186400 | DOI:10.1111/aos.17495

Mol Biotechnol. 2025 Apr 5. doi: 10.1007/s12033-025-01417-5. Online ahead of print.

ABSTRACT

Viral diseases pose a significant threat to public health, highlighting the importance of understanding protein-protein interactions between hosts and viruses for therapeutic development. However, this process is often expensive and time-consuming, especially given the rapid evolution of viruses. Machine learning algorithms and artificial intelligence have emerged as powerful tools for efficiently identifying these interactions. This study aims to develop a machine learning-based model to predict protein interactions between viral pathogens and human hosts while analyzing the factors influencing these interactions. The prediction model was constructed using three machine learning algorithms: Random Forest (RF), XGBoost (XGB), and Artificial Neural Networks (ANN). Each algorithm underwent rigorous testing. The modeling features included physicochemical properties, motifs, and amino acid sequences. Model performance was evaluated using fitness, accuracy, precision, sensitivity, and specificity metrics, with validation conducted via the K-fold method. The accuracy of the RF, XGB, and ANN models was 87%, 86%, and 86%, respectively. By integrating dimensionality reduction and clustering techniques, the accuracy of the RF model improved to 90%. Traditionally, studying host-pathogen interactions is labor intensive and costly. The integration of machine learning algorithms into this field significantly enhances the efficiency of analyzing viral pathogen-human host interactions. This study demonstrates the effectiveness of such an approach and provides valuable insights for future research. The results are accessible to researchers through a web application at http://vhi.sysbiomed.ir .

PMID:40186829 | DOI:10.1007/s12033-025-01417-5

Mol Ther. 2025 Apr 3:S1525-0016(25)00275-8. doi: 10.1016/j.ymthe.2025.04.001. Online ahead of print.

ABSTRACT

Multi-modal datasets have grown exponentially in the last decade. This has created an enormous demand for machine learning models that can predict complex outcomes by leveraging cellular, molecular and humoral profiles. Corresponding inference of mechanisms can help uncover new therapeutic targets. Here, we discuss how biological principles guide the design of predictive models and how interpretable machine learning can lead to novel mechanistic insights. We provide descriptions of multiple learning techniques and how suited they are to domain adaptations. Finally, we talk about broad learning capabilities of foundation models on large datasets and whether they can be used to provide meaningful inference about biological datasets.

PMID:40186352 | DOI:10.1016/j.ymthe.2025.04.001

Gut Microbes. 2025 Dec;17(1):2489074. doi: 10.1080/19490976.2025.2489074. Epub 2025 Apr 4.

ABSTRACT

The human gut microbiome, crucial in various diseases, can be utilized to develop diagnostic models through machine learning (ML). The specific tools and parameters used in model construction such as data preprocessing, batch effect removal and modeling algorithms can impact model performance and generalizability. To establish an generally applicable workflow, we divided the ML process into three above-mentioned steps and optimized each sequentially using 83 gut microbiome cohorts across 20 diseases. We tested a total of 156 tool-parameter-algorithm combinations and benchmarked them according to internal- and external- AUCs. At the data preprocessing step, we identified four data preprocessing methods that performed well for regression-type algorithms and one method that excelled for non-regression-type algorithms. At the batch effect removal step, we identified the "ComBat" function from the sva R package as an effective batch effect removal method and compared the performance of various algorithms. Finally, at the ML algorithm selection step, we found that Ridge and Random Forest ranked the best. Our optimized work flow performed similarly comparing with previous exhaustive methods for disease-specific optimizations, thus is generally applicable and can provide a comprehensive guideline for constructing diagnostic models for a range of diseases, potentially serving as a powerful tool for future medical diagnostics.

PMID:40186338 | DOI:10.1080/19490976.2025.2489074