Front Psychol. 2024 Dec 24;15:1509392. doi: 10.3389/fpsyg.2024.1509392. eCollection 2024.

ABSTRACT

INTRODUCTION: While the fact that visual stimuli synthesized by Artificial Neural Networks (ANN) may evoke emotional reactions is documented, the precise mechanisms that connect the strength and type of such reactions with the ways of how ANNs are used to synthesize visual stimuli are yet to be discovered. Understanding these mechanisms allows for designing methods that synthesize images attenuating or enhancing selected emotional states, which may provide unobtrusive and widely-applicable treatment of mental dysfunctions and disorders.

METHODS: The Convolutional Neural Network (CNN), a type of ANN used in computer vision tasks which models the ways humans solve visual tasks, was applied to synthesize ("dream" or "hallucinate") images with no semantic content to maximize activations of neurons in precisely-selected layers in the CNN. The evoked emotions of 150 human subjects observing these images were self-reported on a two-dimensional scale (arousal and valence) utilizing self-assessment manikin (SAM) figures. Correlations between arousal and valence values and image visual properties (e.g., color, brightness, clutter feature congestion, and clutter sub-band entropy) as well as the position of the CNN's layers stimulated to obtain a given image were calculated.

RESULTS: Synthesized images that maximized activations of some of the CNN layers led to significantly higher or lower arousal and valence levels compared to average subject's reactions. Multiple linear regression analysis found that a small set of selected image global visual features (hue, feature congestion, and sub-band entropy) are significant predictors of the measured arousal, however no statistically significant dependencies were found between image global visual features and the measured valence.

CONCLUSION: This study demonstrates that the specific method of synthesizing images by maximizing small and precisely-selected parts of the CNN used in this work may lead to synthesis of visual stimuli that enhance or attenuate emotional reactions. This method paves the way for developing tools that stimulate, in a non-invasive way, to support wellbeing (manage stress, enhance mood) and to assist patients with certain mental conditions by complementing traditional methods of therapeutic interventions.

PMID:39776961 | PMC:PMC11703666 | DOI:10.3389/fpsyg.2024.1509392

Patterns (N Y). 2024 Dec 9;5(12):101116. doi: 10.1016/j.patter.2024.101116. eCollection 2024 Dec 13.

ABSTRACT

To achieve adequate trust in patient-critical medical tasks, artificial intelligence must be able to recognize instances where they cannot operate confidently. Ensemble methods are deployed to estimate uncertainty, but models in an ensemble often share the same vulnerabilities to adversarial attacks. We propose an ensemble approach based on feature decorrelation and Fourier partitioning for teaching networks diverse features, reducing the chance of perturbation-based fooling. We test our approach against white-box attacks in single- and multi-channel electrocardiogram classification and adapt adversarial training and DVERGE into an ensemble framework for comparison. Our results indicate that the combination of decorrelation and Fourier partitioning maintains performance on unperturbed data while demonstrating superior uncertainty estimation on projected gradient descent and smooth adversarial attacks of various magnitudes. Furthermore, our approach does not require expensive optimization with adversarial samples during training. These methods can be applied to other tasks for more robust models.

PMID:39776851 | PMC:PMC11701855 | DOI:10.1016/j.patter.2024.101116

Infect Drug Resist. 2025 Jan 3;18:31-42. doi: 10.2147/IDR.S482584. eCollection 2025.

ABSTRACT

BACKGROUND: Early differentiation between spinal tuberculosis (STB) and acute osteoporotic vertebral compression fracture (OVCF) is crucial for determining the appropriate clinical management and treatment pathway, thereby significantly impacting patient outcomes.

OBJECTIVE: To evaluate the efficacy of deep learning (DL) models using reconstructed sagittal CT images in the differentiation of early STB from acute OVCF, with the aim of enhancing diagnostic precision, reducing reliance on MRI and biopsies, and minimizing the risks of misdiagnosis.

METHODS: Data were collected from 373 patients, with 302 patients recruited from a university-affiliated hospital serving as the training and internal validation sets, and an additional 71 patients from another university-affiliated hospital serving as the external validation set. MVITV2, Efficient-Net-B5, ResNet101, and ResNet50 were used as the backbone networks for DL model development, training, and validation. Model evaluation was based on accuracy, precision, sensitivity, F1 score, and area under the curve (AUC). The performance of the DL models was compared with the diagnostic accuracy of two spine surgeons who performed a blinded review.

RESULTS: The MVITV2 model outperformed other architectures in the internal validation set, achieving accuracy of 98.98%, precision of 100%, sensitivity of 97.97%, F1 score of 98.98%, and AUC of 0.997. The performance of the DL models notably exceeded that of the spine surgeons, who achieved accuracy rates of 77.38% and 93.56%. The external validation confirmed the models' robustness and generalizability.

CONCLUSION: The DL models significantly improved the differentiation between STB and OVCF, surpassing experienced spine surgeons in diagnostic accuracy. These models offer a promising alternative to traditional imaging and invasive procedures, potentially promoting early and accurate diagnosis, reducing healthcare costs, and improving patient outcomes. The findings underscore the potential of artificial intelligence for revolutionizing spinal disease diagnostics, and have substantial clinical implications.

PMID:39776757 | PMC:PMC11706012 | DOI:10.2147/IDR.S482584

Clin Med Insights Oncol. 2025 Jan 6;19:11795549241311408. doi: 10.1177/11795549241311408. eCollection 2025.

ABSTRACT

Despite the expanding therapeutic options available to cancer patients, therapeutic resistance, disease recurrence, and metastasis persist as hallmark challenges in the treatment of cancer. The rise to prominence of generative artificial intelligence (GenAI) in many realms of human activities is compelling the consideration of its capabilities as a potential lever to advance the development of effective cancer treatments. This article presents a hypothetical case study on the application of generative pre-trained transformers (GPTs) to the treatment of metastatic prostate cancer (mPC). The case explores the design of GPT-supported adaptive intermittent therapy for mPC. Testosterone and prostate-specific antigen (PSA) are assumed to be repeatedly monitored while treatment may involve a combination of androgen deprivation therapy (ADT), androgen receptor-signalling inhibitors (ARSI), chemotherapy, and radiotherapy. The analysis covers various questions relevant to the configuration, training, and inferencing of GPTs for the case of mPC treatment with a particular attention to risk mitigation regarding the hallucination problem and its implications to clinical integration of GenAI technologies. The case study provides elements of an actionable pathway to the realization of GenAI-assisted adaptive treatment of metastatic prostate cancer. As such, the study is expected to help facilitate the design of clinical trials of GenAI-supported cancer treatments.

PMID:39776668 | PMC:PMC11701910 | DOI:10.1177/11795549241311408

J Med Imaging (Bellingham). 2025 Jan;12(1):014501. doi: 10.1117/1.JMI.12.1.014501. Epub 2025 Jan 6.

ABSTRACT

PURPOSE: The prevalence of type 2 diabetes mellitus (T2DM) has been steadily increasing over the years. We aim to predict the occurrence of T2DM using mammography images within 5 years using two different methods and compare their performance.

APPROACH: We examined 312 samples, including 110 positive cases (developed T2DM after 5 years) and 202 negative cases (did not develop T2DM) using two different methods. In the first method, a radiomics-based approach, we utilized radiomics features and machine learning (ML) algorithms. The entire breast region was chosen as the region of interest for extracting radiomics features. Then, a binary breast image was created from which we extracted 668 features and analyzed them using various ML algorithms. In the second method, a complex convolutional neural network (CNN) with a modified ResNet architecture and various kernel sizes was applied to raw mammography images for the prediction task. A nested, stratified five-fold cross-validation was done for both parts A and B to compute accuracy, sensitivity, specificity, and area under the receiver operating curve (AUROC). Hyperparameter tuning was also done to enhance the model's performance and reliability.

RESULTS: The radiomics approach's light gradient boosting model gave 68.9% accuracy, 30.7% sensitivity, 89.5% specificity, and 0.63 AUROC. The CNN method achieved an AUROC of 0.58 over 20 epochs.

CONCLUSION: Radiomics outperformed CNN by 0.05 in terms of AUROC. This may be due to the more straightforward interpretability and clinical relevance of predefined radiomics features compared with the complex, abstract features learned by CNNs.

PMID:39776665 | PMC:PMC11702674 | DOI:10.1117/1.JMI.12.1.014501

Biol Imaging. 2024 Nov 15;4:e13. doi: 10.1017/S2633903X24000096. eCollection 2024.

ABSTRACT

We propose a neural network architecture and a training procedure to estimate blurring operators and deblur images from a single degraded image. Our key assumption is that the forward operators can be parameterized by a low-dimensional vector. The models we consider include a description of the point spread function with Zernike polynomials in the pupil plane or product-convolution expansions, which incorporate space-varying operators. Numerical experiments show that the proposed method can accurately and robustly recover the blur parameters even for large noise levels. For a convolution model, the average signal-to-noise ratio of the recovered point spread function ranges from 13 dB in the noiseless regime to 8 dB in the high-noise regime. In comparison, the tested alternatives yield negative values. This operator estimate can then be used as an input for an unrolled neural network to deblur the image. Quantitative experiments on synthetic data demonstrate that this method outperforms other commonly used methods both perceptually and in terms of SSIM. The algorithm can process a 512 512 image under a second on a consumer graphics card and does not require any human interaction once the operator parameterization has been set up.1.

PMID:39776610 | PMC:PMC11704139 | DOI:10.1017/S2633903X24000096

Biol Imaging. 2024 Dec 20;4:e16. doi: 10.1017/S2633903X24000151. eCollection 2024.

ABSTRACT

With the fast development of modern microscopes and bioimaging techniques, an unprecedentedly large amount of imaging data is being generated, stored, analyzed, and shared through networks. The size of the data poses great challenges for current data infrastructure. One common way to reduce the data size is by image compression. This study analyzes multiple classic and deep-learning-based image compression methods, as well as an empirical study on their impact on downstream deep-learning-based image processing models. We used deep-learning-based label-free prediction models (i.e., predicting fluorescent images from bright-field images) as an example downstream task for the comparison and analysis of the impact of image compression. Different compression techniques are compared in compression ratio, image similarity, and, most importantly, the prediction accuracy of label-free models on original and compressed images. We found that artificial intelligence (AI)-based compression techniques largely outperform the classic ones with minimal influence on the downstream 2D label-free tasks. In the end, we hope this study could shed light on the potential of deep-learning-based image compression and raise the awareness of the potential impacts of image compression on downstream deep-learning models for analysis.

PMID:39776609 | PMC:PMC11704128 | DOI:10.1017/S2633903X24000151

Biol Imaging. 2024 Nov 22;4:e14. doi: 10.1017/S2633903X24000114. eCollection 2024.

ABSTRACT

This manuscript showcases the latest advancements in deepImageJ, a pivotal Fiji/ImageJ plugin for bioimage analysis in life sciences. The plugin, known for its user-friendly interface, facilitates the application of diverse pre-trained convolutional neural networks to custom data. The manuscript demonstrates several deepImageJ capabilities, particularly in deploying complex pipelines, three-dimensional (3D) image analysis, and processing large images. A key development is the integration of the Java Deep Learning Library, expanding deepImageJ's compatibility with various deep learning (DL) frameworks, including TensorFlow, PyTorch, and ONNX. This allows for running multiple engines within a single Fiji/ImageJ instance, streamlining complex bioimage analysis workflows. The manuscript details three case studies to demonstrate these capabilities. The first case study explores integrated image-to-image translation followed by nuclei segmentation. The second case study focuses on 3D nuclei segmentation. The third case study showcases large image volume segmentation and compatibility with the BioImage Model Zoo. These use cases underscore deepImageJ's versatility and power to make advanced DLmore accessible and efficient for bioimage analysis. The new developments within deepImageJ seek to provide a more flexible and enriched user-friendly framework to enable next-generation image processing in life science.

PMID:39776608 | PMC:PMC11704127 | DOI:10.1017/S2633903X24000114

Mach Learn Clin Neuroimaging (2024). 2025;15266:13-23. doi: 10.1007/978-3-031-78761-4_2. Epub 2024 Dec 6.

ABSTRACT

Quantitative Susceptibility Mapping (QSM) is a technique that derives tissue magnetic susceptibility distributions from phase measurements obtained through Magnetic Resonance (MR) imaging. This involves solving an ill-posed dipole inversion problem, however, and thus time-consuming and cumbersome data acquisition from several distinct head orientations becomes necessary to obtain an accurate solution. Most recent (supervised) deep learning methods for single-phase QSM require training data obtained via multiple orientations. In this work, we present an alternative unsupervised learning approach that can efficiently train on single-orientation measurement data alone, named ProxiMO (Proximal Multi-Operator), combining Learned Proximal Convolutional Neural Networks (LP-CNN) with multi-operator imaging (MOI). This integration enables LP-CNN training for QSM on single-phase data without ground truth reconstructions. We further introduce a semi-supervised variant, which further boosts the reconstruction performance, compared to the traditional supervised fashions. Extensive experiments on multicenter datasets illustrate the advantage of unsupervised training and the superiority of the proposed approach for QSM reconstruction. Code is available at https://github.com/shmuelor/ProxiMO.

PMID:39776602 | PMC:PMC11705005 | DOI:10.1007/978-3-031-78761-4_2

Sci Transl Med. 2025 Jan 8;17(780):eadk8623. doi: 10.1126/scitranslmed.adk8623. Epub 2025 Jan 8.

ABSTRACT

Idiopathic pulmonary fibrosis (IPF) is a progressive interstitial lung disease in which repetitive epithelial injury and incomplete alveolar repair result in accumulation of profibrotic intermediate/transitional "aberrant" epithelial cell states. The mechanisms leading to the emergence and persistence of aberrant epithelial populations in the distal lung remain incompletely understood. By interrogating single-cell RNA sequencing (scRNA-seq) data from patients with IPF and a mouse model of repeated lung epithelial injury, we identified persistent activation of hypoxia-inducible factor (HIF) signaling in these aberrant epithelial cells. Using mouse genetic lineage-tracing strategies together with scRNA-seq, we found that these disease-emergent aberrant epithelial cells predominantly arose from airway-derived (Scgb1a1-CreER-traced) progenitors and exhibited transcriptional programs of Hif2a activation. In mice treated with repetitive intratracheal bleomycin, deletion of Epas1 (Hif2a) but not Hif1a, from airway-derived progenitors, or administration of the small-molecule HIF2 inhibitor PT-2385, using both prevention and rescue approaches, attenuated experimental lung fibrosis, reduced the appearance of aberrant epithelial cells, and promoted alveolar repair. In mouse alveolar organoids, genetic or pharmacologic inhibition of Hif2 promoted alveolar differentiation of airway-derived epithelial progenitors. In addition, treatment of human distal lung organoids with PT-2385 increased colony-forming efficiency, enhanced protein and transcriptional markers of alveolar type 2 epithelial cell maturation, and prevented the emergence of aberrant epithelial cells. Together, these studies showed that HIF2 activation drives the emergence of aberrant epithelial populations after repetitive injury and that targeted HIF2 inhibition may represent an effective therapeutic strategy to promote functional alveolar repair in IPF and other interstitial lung diseases.

PMID:39772774 | DOI:10.1126/scitranslmed.adk8623

Patterns (N Y). 2024 Dec 6;5(12):101115. doi: 10.1016/j.patter.2024.101115. eCollection 2024 Dec 13.

ABSTRACT

Guidelines in statistical modeling for genomics hold that simpler models have advantages over more complex ones. Potential advantages include cost, interpretability, and improved generalization across datasets or biological contexts. We directly tested the assumption that small gene signatures generalize better by examining the generalization of mutation status prediction models across datasets (from cell lines to human tumors and vice versa) and biological contexts (holding out entire cancer types from pan-cancer data). We compared model selection between solely cross-validation performance and combining cross-validation performance with regularization strength. We did not observe that more regularized signatures generalized better. This result held across both generalization problems and for both linear models (LASSO logistic regression) and non-linear ones (neural networks). When the goal of an analysis is to produce generalizable predictive models, we recommend choosing the ones that perform best on held-out data or in cross-validation instead of those that are smaller or more regularized.

PMID:39776849 | PMC:PMC11701843 | DOI:10.1016/j.patter.2024.101115

Theranostics. 2025 Jan 1;15(3):1094-1109. doi: 10.7150/thno.96488. eCollection 2025.

ABSTRACT

Rationale: The mannose receptor (CD206, expressed by the gene Mrc1) is a surface marker overexpressed by anti-inflammatory and pro-tumoral macrophages. As such, CD206+ macrophages play key roles in the immune response to different pathophysiological conditions and represent a promising diagnostic and therapeutic target. However, methods to specifically target these cells remain challenging. In this study, we describe a multi-mannose approach to develop CD206-targeting fluorescent and MRI agents that specifically and sensitively detect and monitor CD206+ macrophage immune response in different disease conditions. Methods: We designed and synthesized fluorescent agents MR1-cy5 and MR2-cy5, and MRI agents Mann2-DTPA-Gd and MannGdFish. Cellular assays using pro-inflammatory and anti-inflammatory macrophages differentiated from RAW 264.7 cells were performed, and signals were detected by fluorescence microscopy and inductively coupled plasma mass spectrometry (ICP-MS) to validate specificity in vitro. In vivo specificity and efficacy of the agents were evaluated by MRI in a subcutaneous wound healing model and experimental glioma with Mrc1 +/+ without and with D-mannose treatment, Mrc1 +/-, and Mrc1 -/- mice, and in stroke. One-way ANOVA and two-way ANOVA tests were used for data analysis. P < 0.05 was considered statistically different. Results: Both in vitro fluorescence imaging with MR2-cy5, ICP-MS with Mann2-DTPA-Gd, and in vivo MRI in Mrc1 -/- mice confirmed the specificity of our approach. Mann2-DTPA-Gd MRI can track the changes of CD206+ macrophages at different stages of wound healing, correlating well with flow cytometry data using another anti-inflammatory macrophage marker (arginase-1). The specificity and efficacy of Mann2-DPTA-Gd were further validated in experimental glioma, in which Mann2-DTPA-Gd imaging detected CD206+ tumor-associated macrophages (TAMs), demonstrated significantly decreased signals in Mrc1 +/- mice and Mrc1 -/- mice, and tracked treatment changes in D-mannose-treated Mrc1 +/+ mice. Furthermore, Mann2-DTPA-Gd can report microglia/macrophages and correlate with histology in stroke. The more Gd-stable agent MannGdFish demonstrated similar efficacy as Mann2-DTPA-Gd in vivo with favorable biodistribution and pharmacokinetics. Conclusion: We have developed a fluorescent agent (MR2-cy5) and MRI agents (Mann2-DTPA-Gd and MannGdFish) with two mannose moieties that are highly specific to CD206 and can track CD206+ macrophages in disease models of wound healing, tumor, and neurological disease. Importantly, MannGdFish, with its high specificity, stability, favorable biodistribution, and pharmacokinetics, is a promising translational candidate to noninvasively monitor CD206+ macrophages in repair/regeneration and tumors in patients. In addition, with the specific binding motif to CD206, other imaging modalities and therapeutic agents could also be introduced for theranostic applications.

PMID:39776805 | PMC:PMC11700851 | DOI:10.7150/thno.96488

JBMR Plus. 2024 Dec 10;9(2):ziae159. doi: 10.1093/jbmrpl/ziae159. eCollection 2025 Feb.

ABSTRACT

Human genetic studies have nominated cadherin-like and PC-esterase domain-containing 1 (CPED1) as a candidate target gene mediating bone mineral density (BMD) and fracture risk heritability. Recent efforts to define the role of CPED1 in bone in mouse and human models have revealed complex alternative splicing and inconsistent results arising from gene targeting, making its function in bone difficult to interpret. To better understand the role of CPED1 in adult bone mass and morphology, we conducted a comprehensive genetic and phenotypic analysis of cped1 in zebrafish, an emerging model for bone and mineral research. We analyzed two different cped1 mutant lines and performed deep phenotyping to characterize more than 200 measures of adult vertebral, craniofacial, and lean tissue morphology. We also examined alternative splicing of zebrafish cped1 and gene expression in various cell/tissue types. Our studies fail to support an essential role of cped1 in adult zebrafish bone. Specifically, homozygous mutants for both cped1 mutant alleles, which are expected to result in loss-of-function and impact all cped1 isoforms, exhibited no significant differences in the measures examined when compared to their respective wildtype controls, suggesting that cped1 does not significantly contribute to these traits. We identified sequence differences in critical residues of the catalytic triad between the zebrafish and mouse orthologs of CPED1, suggesting that differences in key residues, as well as distinct alternative splicing, could underlie different functions of CPED1 orthologs in the two species. Our studies fail to support a requirement of cped1 in zebrafish bone and lean tissue, adding to evidence that variants at 7q31.31 can act independently of CPED1 to influence BMD and fracture risk.

PMID:39776615 | PMC:PMC11701521 | DOI:10.1093/jbmrpl/ziae159

Gut Microbiome (Camb). 2024 May 31;5:e7. doi: 10.1017/gmb.2024.5. eCollection 2024.

ABSTRACT

Breastfeeding represents a strong selective factor for shaping the infant gut microbiota. Besides providing nutritional requirements for the infant, human milk is a key source of oligosaccharides, human milk oligosaccharides (HMOs), and diverse microbes in early life. This study aimed to evaluate the influence of human milk microbiota and oligosaccharides on the composition of infant faecal microbiota at one, three, and nine months postpartum. We profiled milk microbiota, HMOs, and infant faecal microbiota from 23 mother-infant pairs at these time points. The predominant genera in milk samples were Streptococcus, Staphylococcus, and an unclassified Enterobacteriaceae genus-level taxon (Enterobacteriaceae uncl.), whereas the infant faecal microbiota was predominated by Bifidobacterium, Bacteroides, and Enterobacteriaceae uncl. Mother-infant dyads frequently shared bacterial amplicon sequence variants (ASVs) belonging to the genera Bifidobacterium, Streptococcus, Enterobacteriaceae uncl., Veillonella, Bacteroides, and Haemophilus. The individual HMO concentrations in the milk showed either no change or decreased over the lactation period, except for 3-fucosyllactose (3-FL), which increased. Neither maternal secretor status nor HMO concentrations were significantly associated with microbiota composition at the different ages or the bacterial ASVs of maternal milk and infant faeces. This study suggests an age-dependent role of milk microbes in shaping the gut microbiota, while variations in HMO concentrations show limited influence.

PMID:39776540 | PMC:PMC11706684 | DOI:10.1017/gmb.2024.5

Lab Chip. 2025 Jan 8. doi: 10.1039/d4lc00779d. Online ahead of print.

ABSTRACT

Point-of-care testing (POCT) holds significant importance in the field of infectious disease prevention and control, as well as personalized precision medicine. The emerging microfluidics, capable of minimal reagent consumption, integration, and a high degree of automation, play a pivotal role in POCT. Centrifugal microfluidics, also termed lab-on-a-disc (LOAD), is a significant subfield of microfluidics that integrates crucial analytical steps onto a single chip, thereby optimizing the process and enabling high-throughput, automated analysis. By utilizing rotational mechanics to precisely control fluid dynamics without external pressure sources, centrifugal microfluidics facilitates swift operations ideal for urgent medical and field settings. This review provides a comprehensive overview of the latest advancements in centrifugal microfluidics for POCT, covering both theoretical principles and practical applications. We begin by introducing the fundamental operational principles, fluidic control mechanisms, and signal output detection methods. Subsequently, we delve into the typical applications of centrifugal microfluidic platforms in immunoassays, nucleic acid testing, antimicrobial susceptibility testing, and other tests. We also discuss the strengths and potential limitations of centrifugal microfluidic platforms, underscoring their transformative impact on traditional conventional procedures and their significant role in diagnostic practices.

PMID:39776118 | DOI:10.1039/d4lc00779d

Naturwissenschaften. 2025 Jan 8;112(1):5. doi: 10.1007/s00114-024-01950-4.

ABSTRACT

Physiological or genetic assays and computational modeling are valuable tools for understanding animals' visual discrimination capabilities. Yet sometimes, the results generated by these methods appear not to jive with other aspects of an animal's appearance or natural history, and behavioral confirmatory tests are warranted. Here we examine the peculiar case of a male jumping spider that displays red, black, white, and UV color patches during courtship despite the fact that, according to microspectrophotometry and color vision modeling, they are unlikely able to discriminate red from black. To test whether some optical or neurological component could have been missed using these methods, we conduct mate choice experiments. Some females are presented with a choice between males with their red leg coloration painted over with either red or black paint, while other females are presented with a choice between males with the same coloration painted over by either red or white paint. This latter pairing of red and white males should have been easily distinguishable to the spiders and served as a control to ensure our experimental setup was conducive to natural mating behavior. Red males were more likely to mate than white males (P = 0.035), whereas red and black males had identical mating success (P = 1.0). This suggests that previous physiological and computational work on these spiders was correct in concluding that they are unable to discriminate between red and black. Any functional significance of displaying both colors, rather than only black, remains unresolved.

PMID:39775916 | DOI:10.1007/s00114-024-01950-4

Bioinformatics. 2025 Jan 7:btaf005. doi: 10.1093/bioinformatics/btaf005. Online ahead of print.

ABSTRACT

MOTIVATION: The identification of biologically meaningful domains is a central step in the analysis of spatial transcriptomic data.

RESULTS: Following Occam's razor, we show that a simple PCA-based algorithm for unsupervised spatial domain identification rivals the performance of ten competing state-of-the-art methods across six single-cell spatial transcriptomic datasets. Our reductionist approach, NichePCA, provides researchers with intuitive domain interpretation and excels in execution speed, robustness, and scalability.

AVAILABILITY AND IMPLEMENTATION: The code is available at https://github.com/imsb-uke/nichepca.

SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

PMID:39775801 | DOI:10.1093/bioinformatics/btaf005

Int J Comput Assist Radiol Surg. 2025 Jan 7. doi: 10.1007/s11548-024-03310-z. Online ahead of print.

ABSTRACT

PURPOSE: Lung fissure segmentation on CT images often relies on 3D convolutional neural networks (CNNs). However, 3D-CNNs are inefficient for detecting thin structures like the fissures, which make up a tiny fraction of the entire image volume. We propose to make lung fissure segmentation more efficient by using geometric deep learning (GDL) on sparse point clouds.

METHODS: We abstract image data with sparse keypoint (KP) clouds. We train GDL models to segment the point cloud, comparing three major paradigms of models (PointNets, graph convolutional networks (GCNs), and PointTransformers). From the sparse point segmentations, 3D meshes of the objects are reconstructed to obtain a dense surface. The state-of-the-art Poisson surface reconstruction (PSR) makes up most of the time in our pipeline. Therefore, we propose an efficient point cloud to mesh autoencoder (PC-AE) that deforms a template mesh to fit a point cloud in a single forward pass. Our pipeline is evaluated extensively and compared to the 3D-CNN gold standard nnU-Net on diverse clinical and pathological data.

RESULTS: GCNs yield the best trade-off between inference time and accuracy, being 21 × faster with only 1.4 × increased error over the nnU-Net. Our PC-AE also achieves a favorable trade-off, being 3 × faster at 1.5 × the error compared to the PSR.

CONCLUSION: We present a KP-based fissure segmentation pipeline that is more efficient than 3D-CNNs and can greatly speed up large-scale analyses. A novel PC-AE for efficient mesh reconstruction from sparse point clouds is introduced, showing promise not only for fissure segmentation. Source code is available on https://github.com/kaftanski/fissure-segmentation-IJCARS.

PMID:39775630 | DOI:10.1007/s11548-024-03310-z

Eur J Clin Microbiol Infect Dis. 2025 Jan 7. doi: 10.1007/s10096-024-05030-3. Online ahead of print.

ABSTRACT

PURPOSE: We designed and tested a point of care test panel to detect E.coli and antibiotic susceptibility in urine samples from patients at the point of care in the urological department. The aim of this approach is to facilitate choosing an appropriate antibiotic for urinary tract infections (UTI) at first presentation in the context of increasing antibiotic resistance in uropathogens worldwide.

METHODS: We analyzed 162 E.coli isolates from samples from a university urological department to determine phenotypic and genotypic resistance data. With this data we created customized LAMP (loop-mediated isothermal amplification) panels for a commercial machine with which to detect and possibly quantify E.coli and six antibiotic resistance determinants. In a second step we tested these panel(s) for diagnostic accuracy on 1596 urine samples and compared with routine microbiological culture.

RESULTS: E.coli was detected with 95.4% sensitivity and 96.1% specificity. Dynamics of the LAMP amplification could be used to gauge bacterial loads in the samples. Antibiotic sensitivity was detected with good negative (sensitive) predictive values: ampicillin 92.8%, ampicillin/sulbactam 96.4%, cefuroxime 92.8%, cefotaxime 97.8%, trimethoprim/sulfamethoxazole 96.5%, ciprofloxacin 96.8%.

CONCLUSION: The LAMP panel provided E.coli detection and sensitivity information within one hour and thus could principally guide initial antibiotic therapy upon patients presenting with UTI. The panel helps to select initial adequate antibiotic therapy as well as providing diagnostic stewardship. Follow up investigations will expand the test system to other uropathogens.

PMID:39775368 | DOI:10.1007/s10096-024-05030-3

PLoS Comput Biol. 2024 Dec 30;20(12):e1012653. doi: 10.1371/journal.pcbi.1012653. Online ahead of print.

ABSTRACT

Phylogenies depicting the evolutionary history of genetically heterogeneous subpopulations of cells from the same cancer, i.e., cancer phylogenies, offer valuable insights about cancer development and guide treatment strategies. Many methods exist that reconstruct cancer phylogenies using point mutations detected with bulk DNA sequencing. However, these methods become inaccurate when reconstructing phylogenies with more than 30 mutations, or, in some cases, fail to recover a phylogeny altogether. Here, we introduce Orchard, a cancer phylogeny reconstruction algorithm that is fast and accurate using up to 1000 mutations. Orchard samples without replacement from a factorized approximation of the posterior distribution over phylogenies, a novel result derived in this paper. Each factor in this approximate posterior corresponds to a conditional distribution for adding a new mutation to a partially built phylogeny. Orchard optimizes each factor sequentially, generating a sequence of incrementally larger phylogenies that ultimately culminate in a complete tree containing all mutations. Our evaluations demonstrate that Orchard outperforms state-of-the-art cancer phylogeny reconstruction methods in reconstructing more plausible phylogenies across 90 simulated cancers and 14 B-progenitor acute lymphoblastic leukemias (B-ALLs). Remarkably, Orchard accurately reconstructs cancer phylogenies using up to 1,000 mutations. Additionally, we demonstrate that the large and accurate phylogenies reconstructed by Orchard are useful for identifying patterns of somatic mutations and genetic variations among distinct cancer cell subpopulations.

PMID:39775053 | DOI:10.1371/journal.pcbi.1012653