RESUMEN

The use of prior knowledge in the machine learning framework has been considered a potential tool to handle the curse of dimensionality in genetic and genomics data. Although random forest (RF) represents a flexible non-parametric approach with several advantages, it can provide poor accuracy in high-dimensional settings, mainly in scenarios with small sample sizes. We propose a knowledge-slanted RF that integrates biological networks as prior knowledge into the model to improve its performance and explainability, exemplifying its use for selecting and identifying relevant genes. knowledge-slanted RF is a combination of two stages. First, prior knowledge represented by graphs is translated by running a random walk with restart algorithm to determine the relevance of each gene based on its connection and localization on a protein-protein interaction network. Then, each relevance is used to modify the selection probability to draw a gene as a candidate split-feature in the conventional RF. Experiments in simulated datasets with very small sample sizes ( n ≤ 30 ) comparing knowledge-slanted RF against conventional RF and logistic lasso regression, suggest an improved precision in outcome prediction compared to the other methods. The knowledge-slanted RF was completed with the introduction of a modified version of the Boruta feature selection algorithm. Finally, knowledge-slanted RF identified more relevant biological genes, offering a higher level of explainability for users than conventional RF. These findings were corroborated in one real case to identify relevant genes to calcific aortic valve stenosis.

RESUMEN

The precise extraction of winter wheat planting structure holds significant importance for food security risk assessment, agricultural resource management, and governmental decision-making. This study proposed a method for extracting the winter wheat planting structure by taking into account the growth phenology of winter wheat. Utilizing the fitting effect index, the optimal Savitzky-Golay (S-G) filtering parameter combination was determined automatically to achieve automated filtering and reconstruction of NDVI time series data. The phenological phases of winter wheat growth was identified automatically using a threshold method, and subsequently, a model for extracting the winter wheat planting structure was constructed based on three key phenological stages, including seeding, heading, and harvesting, with the combination of hierarchical classification principles. A priori sample library was constructed using historical data on winter wheat distribution to verify the accuracy of the extracted results. The validation of fitting effect on different surfaces demonstrated that the optimal filtering parameters for S-G filtering could be obtained automatically by using the fitting effect index. The extracted winter wheat phenological phases showed good consistency with ground-based observational results and MOD12Q2 phenological products. Validation against statistical yearbook data and the proposed priori knowledge base exhibited high statistical accuracy and spatial precision, with an extracting accuracy of 94.92%, a spatial positioning accuracy of 93.26%, and a kappa coefficient of 0.9228. The results indicated that the proposed method for winter wheat planting structure extracting can identify winter wheat areas rapidly and significantly. Furthermore, this method does not require training samples or manual experience, and exhibits strong transferability.

Asunto(s)

RESUMEN

Background and purpose: Accurate and automated segmentation of targets and organs-at-risk (OARs) is crucial for the successful clinical application of online adaptive radiotherapy (ART). Current methods for cone-beam computed tomography (CBCT) auto-segmentation face challenges, resulting in segmentations often failing to reach clinical acceptability. Current approaches for CBCT auto-segmentation overlook the wealth of information available from initial planning and prior adaptive fractions that could enhance segmentation precision. Materials and methods: We introduce a novel framework that incorporates data from a patient's initial plan and previous adaptive fractions, harnessing this additional temporal context to significantly refine the segmentation accuracy for the current fraction's CBCT images. We present LSTM-UNet, an innovative architecture that integrates Long Short-Term Memory (LSTM) units into the skip connections of the traditional U-Net framework to retain information from previous fractions. The models underwent initial pre-training with simulated data followed by fine-tuning on a clinical dataset. Results: Our proposed model's segmentation predictions yield an average Dice similarity coefficient of 79% from 8 Head & Neck organs and targets, compared to 52% from a baseline model without prior knowledge and 78% from a baseline model with prior knowledge but no memory. Conclusions: Our proposed model excels beyond baseline segmentation frameworks by effectively utilizing information from prior fractions, thus reducing the effort of clinicians to revise the auto-segmentation results. Moreover, it works together with registration-based methods that offer better prior knowledge. Our model holds promise for integration into the online ART workflow, offering precise segmentation capabilities on synthetic CT images.

RESUMEN

A precise radiotherapy plan is crucial to ensure accurate segmentation of glioblastomas (GBMs) for radiation therapy. However, the traditional manual segmentation process is labor-intensive and heavily reliant on the experience of radiation oncologists. In this retrospective study, a novel auto-segmentation method is proposed to address these problems. To assess the method's applicability across diverse scenarios, we conducted its development and evaluation using a cohort of 148 eligible patients drawn from four multicenter datasets and retrospective data collection including noncontrast CT, multisequence MRI scans, and corresponding medical records. All patients were diagnosed with histologically confirmed high-grade glioma (HGG). A deep learning-based method (PKMI-Net) for automatically segmenting gross tumor volume (GTV) and clinical target volumes (CTV1 and CTV2) of GBMs was proposed by leveraging prior knowledge from multimodal imaging. The proposed PKMI-Net demonstrated high accuracy in segmenting, respectively, GTV, CTV1, and CTV2 in an 11-patient test set, achieving Dice similarity coefficients (DSC) of 0.94, 0.95, and 0.92; 95% Hausdorff distances (HD95) of 2.07, 1.18, and 3.95 mm; average surface distances (ASD) of 0.69, 0.39, and 1.17 mm; and relative volume differences (RVD) of 5.50%, 9.68%, and 3.97%. Moreover, the vast majority of GTV, CTV1, and CTV2 produced by PKMI-Net are clinically acceptable and require no revision for clinical practice. In our multicenter evaluation, the PKMI-Net exhibited consistent and robust generalizability across the various datasets, demonstrating its effectiveness in automatically segmenting GBMs. The proposed method using prior knowledge in multimodal imaging can improve the contouring accuracy of GBMs, which holds the potential to improve the quality and efficiency of GBMs' radiotherapy.

RESUMEN

Recent image classification efforts have achieved certain success by incorporating prior information such as labels and logical rules to learn discriminative features. However, these methods overlook the variability of features, resulting in feature inconsistency and fluctuations in model parameter updates, which further contribute to decreased image classification accuracy and model instability. To address this issue, this paper proposes a novel method combining structural prior-driven feature extraction with gradient-momentum (SPGM), from the perspectives of consistent feature learning and precise parameter updates, to enhance the accuracy and stability of image classification. Specifically, SPGM leverages a structural prior-driven feature extraction (SPFE) approach to calculate gradients of multi-level features and original images to construct structural information, which is then transformed into prior knowledge to drive the network to learn features consistent with the original images. Additionally, an optimization strategy integrating gradients and momentum (GMO) is introduced, dynamically adjusting the direction and step size of parameter updates based on the angle and norm of the sum of gradients and momentum, enabling precise model parameter updates. Extensive experiments on CIFAR10 and CIFAR100 datasets demonstrate that the SPGM method significantly reduces the top-1 error rate in image classification, enhances the classification performance, and outperforms state-of-the-art methods.

Asunto(s)

RESUMEN

High school students with better knowledge about back care have fewer problems, but conceptual errors can hinder the acquisition of essential knowledge necessary for developing healthy habits. This study analyzed secondary school students' declarative knowledge and misconceptions related to back care in daily activities. An exploratory cross-sectional study was conducted with 80 girls and 89 boys aged 14-18 years (M = 15.68, SD = 2.12). The Health Questionnaire on Back Care Knowledge in Activities of Daily Living was used to evaluate knowledge using the true answer model (TAM) and the misconception model (MM). Using the test-retest method, both models' reliability was confirmed (TAM = 0.75; MM = 0.77), while only a minimal measurement error was identified (TAM = -0.01; MM = -0.07). The average scores were 6.23 for the TAM and 2.29 for the MM. The results showed no significant differences in both models. The analysis indicated that students had the most accurate knowledge of the location and function of the spine, whereas misconceptions regarding anatomical understanding and body posture usage were common. An analysis of the results under Reassumption Theory emphasizes the significance of comprehending concepts such as load transmission and spinal stability to maintain back health, thus highlighting the need for improved education in these areas to address misconceptions and enhance overall back-care knowledge.

RESUMEN

Recent advances in microfluidics and sequencing technologies allow researchers to explore cellular heterogeneity at single-cell resolution. In recent years, deep learning frameworks, such as generative models, have brought great changes to the analysis of transcriptomic data. Nevertheless, relying on the potential space of these generative models alone is insufficient to generate biological explanations. In addition, most of the previous work based on generative models is limited to shallow neural networks with one to three layers of latent variables, which may limit the capabilities of the models. Here, we propose a deep interpretable generative model called d-scIGM for single-cell data analysis. d-scIGM combines sawtooth connectivity techniques and residual networks, thereby constructing a deep generative framework. In addition, d-scIGM incorporates hierarchical prior knowledge of biological domains to enhance the interpretability of the model. We show that d-scIGM achieves excellent performance in a variety of fundamental tasks, including clustering, visualization, and pseudo-temporal inference. Through topic pathway studies, we found that d-scIGM-learned topics are better enriched for biologically meaningful pathways compared to the baseline models. Furthermore, the analysis of drug response data shows that d-scIGM can capture drug response patterns in large-scale experiments, which provides a promising way to elucidate the underlying biological mechanisms. Lastly, in the melanoma dataset, d-scIGM accurately identified different cell types and revealed multiple melanin-related driver genes and key pathways, which are critical for understanding disease mechanisms and drug development.

Asunto(s)

RESUMEN

Deep learning from collaboration occurs if the learner enacts interactive activities in the sense of leveraging the knowledge externalized by co-learners as resource for own inferencing processes and if these interactive activities in turn promote the learner's deep comprehension outcomes. This experimental study investigates whether inducing dyad members to enact constructive preparation activities can promote deep learning from subsequent collaboration while examining prior knowledge as moderator. In a digital collaborative learning environment, 122 non-expert university students assigned to 61 dyads studied a text about the human circulatory system and then prepared individually for collaboration according to their experimental conditions: the preparation tasks varied across dyads with respect to their generativity, that is, the degree to which they required the learners to enact constructive activities (note-taking, compare-contrast, or explanation). After externalizing their answer to the task, learners in all conditions inspected their partner's externalization and then jointly discussed their text understanding via chat. Results showed that more rather than less generative tasks fostered constructive preparation but not interactive collaboration activities or deep comprehension outcomes. Moderated mediation analyses considering actor and partner effects indicated the indirect effects of constructive preparation activities on deep comprehension outcomes via interactive activities to depend on prior knowledge: when own prior knowledge was relatively low, self-performed but not partner-performed constructive preparation activities were beneficial. When own prior knowledge was relatively high, partner-performed constructive preparation activities were conducive while one's own were ineffective or even detrimental. Given these differential effects, suggestions are made for optimizing the instructional design around generative preparation tasks to streamline the effectiveness of constructive preparation activities for deep learning from digital collaboration.

RESUMEN

Feature binding is the process of integrating features, such as colour and shape, into object representations. A persistent question in the literature concerning whether feature binding is an automatic or resource-demanding process may depend on unitisation, that is, whether the to-be-bound information is intrinsic (belonging to) or extrinsic (contextual). Given extensive evidence showing that Easterners may process information more holistically than Westerners, such cultural differences may be useful to understand the fundamental processes of feature binding in visual working memory (WM). Accordingly, we recruited British and Chinese participants to complete a visual WM task wherein to-be-remembered colours were integrated within (i.e., intrinsic binding) or as backgrounds (i.e., extrinsic binding) of to-be-remembered shapes (Experiments 1 and 2). Experiment 2 further investigated the role of prior knowledge in long-term memory to facilitate feature binding in WM. During retrieval, participants decided among three probes: a target, a lure (i.e., recombination of the presented features), and a new colour/shape. Hierarchical Bayesian multinomial processing tree models were fit to the data to estimate parameters representing binding and item memory. The current results suggest that intrinsic and extrinsic binding memory are similar between the two cultural groups, with no prior knowledge benefits for either intrinsic or extrinsic binding for either cultural group. This result conflicts with the Analytic and Holistic framework and suggests that there are no cultural differences or prior knowledge benefits in feature binding.

RESUMEN

Magnetic particle imaging (MPI) is an emerging non-invasive medical imaging tomography technology based on magnetic particles, with excellent imaging depth penetration, high sensitivity and contrast. Spatial resolution and signal-to-noise ratio (SNR) are key performance metrics for evaluating MPI, which are directly influenced by the gradient of the selection field (SF). Increasing the SF gradient can improve the spatial resolution of MPI, but will lead to a decrease in SNR. Deep learning (DL) methods may enable obtaining high-resolution images from low-resolution images to improve the MPI resolution under low gradient conditions. However, existing DL methods overlook the physical procedures contributing to the blurring of MPI images, resulting in low interpretability and hindering breakthroughs in resolution. To address this issue, we propose a dual-channel end-to-end network with prior knowledge embedding for MPI (DENPK-MPI) to effectively establish a latent mapping between low-gradient and high-gradient images, thus improving MPI resolution without compromising SNR. By seamlessly integrating MPI PSF with DL paradigm, DENPK-MPI leads to a significant improvement in spatial resolution performance. Simulation, phantom, and in vivo MPI experiments have collectively confirmed that our method can improve the resolution of low-gradient MPI images without sacrificing SNR, resulting in a decrease in full width at half maximum by 14.8%-23.8 %, and the accuracy of image reconstruction is 18.2 %-27.3 % higher than other DL methods. In conclusion, we propose a DL method that incorporates MPI prior knowledge, which can improve the spatial resolution of MPI without compromising SNR and possess improved biomedical application.

Asunto(s)

RESUMEN

Radiotherapy is a preferred treatment for brain metastases, which kills cancer cells via high doses of radiation meanwhile hardly avoiding damage to surrounding healthy cells. Therefore, the delineation of organs-at-risk (OARs) is vital in treatment planning to minimize radiation-induced toxicity. However, the following aspects make OAR delineation a challenging task: extremely imbalanced organ sizes, ambiguous boundaries, and complex anatomical structures. To alleviate these challenges, we imitate how specialized clinicians delineate OARs and present a novel cascaded multi-OAR segmentation framework, called OAR-SegNet. OAR-SegNet comprises two distinct levels of segmentation networks: an Anatomical-Prior-Guided network (APG-Net) and a Point-Cloud-Guided network (PCG-Net). Specifically, APG-Net handles segmentation for all organs, where multi-view segmentation modules and a deep prior loss are designed under the guidance of prior knowledge. After APG-Net, PCG-Net refines small organs through the mini-segmentation and the point-cloud alignment heads. The mini-segmentation head is further equipped with the deep prior feature. Extensive experiments were conducted to demonstrate the superior performance of the proposed method compared to other state-of-the-art medical segmentation methods.

Asunto(s)

RESUMEN

Our memories help us plan for the future. In some cases, we use memories to repeat the choices that led to preferable outcomes in the past. The success of these memory-guided decisions depends on close interactions between the hippocampus and medial prefrontal cortex. In other cases, we need to use our memories to deduce hidden connections between the present and past situations to decide the best choice of action based on the expected outcome. Our recent study investigated neural underpinnings of such inferential decisions by monitoring neural activity in the medial prefrontal cortex and hippocampus in rats. We identified several neural activity patterns indicating awake memory trace reactivation and restructuring of functional connectivity among multiple neurons. We also found that these patterns occurred concurrently with the ongoing hippocampal activity when rats recalled past events but not when they planned new adaptive actions. Here, we discussed how these computational properties might contribute to success in inferential decision-making and propose a working model on how the medial prefrontal cortex changes its interaction with the hippocampus depending on whether it reflects on the past or looks into the future.

Asunto(s)

RESUMEN

We introduce a Bayesian approach for biclustering that accounts for the prior functional dependence between genes using hidden Markov models (HMMs). We utilize biological knowledge gathered from gene ontologies and the hidden Markov structure to capture the potential coexpression of neighboring genes. Our interpretable model-based clustering characterized each cluster of samples by three groups of features: overexpressed, underexpressed, and irrelevant features. The proposed methods have been implemented in an R package and are used to analyze both the simulated data and The Cancer Genome Atlas kidney cancer data.

Asunto(s)

RESUMEN

Significant progress in the investigation of how prior knowledge influences episodic memory has been made using three sometimes isolated (but not mutually exclusive) approaches: strictly adult behavioral investigations, computational models, and investigations into the development of the system. Here we point out that these approaches are complementary, each approach informs and is informed by the other. Thus, a natural next step for research is to combine all three approaches to further our understanding of the role of prior knowledge in episodic memory. Here we use studies of memory for expectation-congruent and incongruent information from each of these often disparate approaches to illustrate how combining approaches can be used to test and revise theories from the other. This domain is particularly advantageous because it highlights important features of more general memory processes, further differentiates models of memory, and can shed light on developmental change in the memory system. We then present a case study to illustrate the progress that can be made from integrating all three approaches and highlight the need for more endeavors in this vein. As a first step, we also propose a new computational model of memory that takes into account behavioral and developmental factors that can influence prior knowledge and episodic memory interactions. This integrated approach has great potential for offering novel insights into the relationship between prior knowledge and episodic memory, and cognition more broadly.

RESUMEN

High-throughput technologies are becoming increasingly important in discovering prognostic biomarkers and in identifying novel drug targets. With Mammaprint, Oncotype DX, and many other prognostic molecular signatures breast cancer is one of the paradigmatic examples of the utility of high-throughput data to deliver prognostic biomarkers, that can be represented in a form of a rather short gene list. Such gene lists can be obtained as a set of features (genes) that are important for the decisions of a Machine Learning (ML) method applied to high-dimensional gene expression data. Several studies have identified predictive gene lists for patient prognosis in breast cancer, but these lists are unstable and have only a few genes in common. Instability of feature selection impedes biological interpretability: genes that are relevant for cancer pathology should be members of any predictive gene list obtained for the same clinical type of patients. Stability and interpretability of selected features can be improved by including information on molecular networks in ML methods. Graph Convolutional Neural Network (GCNN) is a contemporary deep learning approach applicable to gene expression data structured by a prior knowledge molecular network. Layer-wise Relevance Propagation (LRP) and SHapley Additive exPlanations (SHAP) are methods to explain individual decisions of deep learning models. We used both GCNN+LRP and GCNN+SHAP techniques to construct feature sets by aggregating individual explanations. We suggest a methodology to systematically and quantitatively analyze the stability, the impact on the classification performance, and the interpretability of the selected feature sets. We used this methodology to compare GCNN+LRP to GCNN+SHAP and to more classical ML-based feature selection approaches. Utilizing a large breast cancer gene expression dataset we show that, while feature selection with SHAP is useful in applications where selected features have to be impactful for classification performance, among all studied methods GCNN+LRP delivers the most stable (reproducible) and interpretable gene lists.

Asunto(s)

RESUMEN

Retinal fundus images serve as a non-invasive modality to obtain information pertaining to retinal vessels through fundus photography, thereby offering insights into cardiovascular and cerebrovascular diseases. Retinal arteriolar morphometry has emerged as the most convenient and fundamental clinical methodology in the realm of patient screening and diagnosis. Nevertheless, the analysis of retinal arterioles is challenging attributable to imaging noise, stochastic fuzzy characteristics, and blurred boundaries proximal to blood vessels. In response to these limitations, we introduce an innovative methodology, named PKSEA-Net, which aims to improve segmentation accuracy by enhancing the perception of edge information in retinal fundus images. PKSEA-Net employs the universal architecture PVT-v2 as the encoder, complemented by a novel decoder architecture consisting of an Edge-Aware Block (EAB) and a Pyramid Feature Fusion Module (PFFM). The EAB block incorporates prior knowledge for supervision and multi-query for multi-task learning, with supervision information derived from an enhanced Full Width at Half Maximum (FWHM) algorithm and gradient map. Moreover, PFFM efficiently integrates multi-scale features through a novel attention fusion method. Additionally, we have collected a Retinal Cross-Sectional Vessel (RCSV) dataset derived from approximately 200 patients in Quzhou People's Hospital to serve as the benchmark dataset. Comparative evaluations with several state-of-the-art (SOTA) networks confirm that PKSEA-Net achieves exceptional experimental performance, thereby establishing its status as a SOTA approach for precise boundary delineation and retinal vessel segmentation.

Asunto(s)

RESUMEN

Introduction: How an event is framed impacts how people judge the morality of those involved, but prior knowledge can influence information processing about an event, which also can impact moral judgments. The current study explored how blame framing and self-reported prior knowledge of a historical act of racial violence, labeled as Riot, Massacre, or Event, impacted individual's cumulative moral judgments regarding the groups involved in the Tulsa Race Massacre (Black Tulsans, the Tulsa Police, and White Tulsans). Methods and results: This study was collected in two cohorts including undergraduates attending the University of Oklahoma and individuals living in the United Kingdom. Participants were randomly assigned to a blame framing condition, read a factual summary of what happened in Tulsa in 1921, and then responded to various moral judgment items about each group. Individuals without prior knowledge had higher average Likert ratings (more blame) toward Black Tulsans and lower average Likert ratings (less blame) toward White Tulsans and the Tulsa Police compared to participants with prior knowledge. This finding was largest when what participants read was framed as a Massacre rather than a Riot or Event. We also found participants with prior knowledge significantly differed in how they made moral judgments across target groups; those with prior knowledge had lower average Likert ratings (less blame) for Black Tulsans and higher average Likert ratings (more blame) for White Tulsans on items pertaining to causal responsibility, intentionality, and punishment compared to participants without prior knowledge. Discussion: Findings suggest that the effect of blame framing on moral judgments is dependent on prior knowledge. Implications for how people interpret both historical and new events involving harmful consequences are discussed.

RESUMEN

Instance segmentation plays an important role in the automatic diagnosis of cervical cancer. Although deep learning-based instance segmentation methods can achieve outstanding performance, they need large amounts of labeled data. This results in a huge consumption of manpower and material resources. To solve this problem, we propose an unsupervised cervical cell instance segmentation method based on human visual simulation, named HVS-Unsup. Our method simulates the process of human cell recognition and incorporates prior knowledge of cervical cells. Specifically, firstly, we utilize prior knowledge to generate three types of pseudo labels for cervical cells. In this way, the unsupervised instance segmentation is transformed to a supervised task. Secondly, we design a Nucleus Enhanced Module (NEM) and a Mask-Assisted Segmentation module (MAS) to address problems of cell overlapping, adhesion, and even scenarios involving visually indistinguishable cases. NEM can accurately locate the nuclei by the nuclei attention feature maps generated by point-level pseudo labels, and MAS can reduce the interference from impurities by updating the weight of the shallow network through the dice loss. Next, we propose a Category-Wise droploss (CW-droploss) to reduce cell omissions in lower-contrast images. Finally, we employ an iterative self-training strategy to rectify mislabeled instances. Experimental results on our dataset MS-cellSeg, the public datasets Cx22 and ISBI2015 demonstrate that HVS-Unsup outperforms existing mainstream unsupervised cervical cell segmentation methods.

Asunto(s)

RESUMEN

BACKGROUND: The automatic generation of radiology reports, which seeks to create a free-text description from a clinical radiograph, is emerging as a pivotal intersection between clinical medicine and artificial intelligence. Leveraging natural language processing technologies can accelerate report creation, enhancing health care quality and standardization. However, most existing studies have not yet fully tapped into the combined potential of advanced language and vision models. OBJECTIVE: The purpose of this study was to explore the integration of pretrained vision-language models into radiology report generation. This would enable the vision-language model to automatically convert clinical images into high-quality textual reports. METHODS: In our research, we introduced a radiology report generation model named ClinicalBLIP, building upon the foundational InstructBLIP model and refining it using clinical image-to-text data sets. A multistage fine-tuning approach via low-rank adaptation was proposed to deepen the semantic comprehension of the visual encoder and the large language model for clinical imagery. Furthermore, prior knowledge was integrated through prompt learning to enhance the precision of the reports generated. Experiments were conducted on both the IU X-RAY and MIMIC-CXR data sets, with ClinicalBLIP compared to several leading methods. RESULTS: Experimental results revealed that ClinicalBLIP obtained superior scores of 0.570/0.365 and 0.534/0.313 on the IU X-RAY/MIMIC-CXR test sets for the Metric for Evaluation of Translation with Explicit Ordering (METEOR) and the Recall-Oriented Understudy for Gisting Evaluation (ROUGE) evaluations, respectively. This performance notably surpasses that of existing state-of-the-art methods. Further evaluations confirmed the effectiveness of the multistage fine-tuning and the integration of prior information, leading to substantial improvements. CONCLUSIONS: The proposed ClinicalBLIP model demonstrated robustness and effectiveness in enhancing clinical radiology report generation, suggesting significant promise for real-world clinical applications.