• Assess the impact of the preprocessing steps (padding, color conversion and resizing) on the performance of deep learning models. • Evaluate the effectiveness of transfer learning, focusing on different pre-trained models and fine-tuning strategies for low-resource and imbalanced datasets. • Propose a robust methodology for recognizing text in non-Latin historical palm leaf manuscripts, addressing the challenges of limited data availability and class imbalance. • Analyse the influence of image quality and class distribution on classification performance. Handwriting recognition (HWR) is currently being adapted to automatically recognize text from digitized historical archives. There has been extensive research on HWR, but studies focusing on old scripts are relatively limited. This is mainly due to script style variations, poor document quality, and a shortage of old script experts. Deep learning algorithms have made significant breakthroughs in the field of HWR. This study evaluates classification tasks with 13 different systems: VGG19, ResNet50V2, ResNet101V2, ResNet152V2, InceptionResNetV2, MobileNetV1, MobileNetV2, NASNetMobile, EfficientNetV2B0, EfficientNetV2B3, EfficientNetV2S, EfficientNetV2M and EfficientNetV2S. Three main observations regarding the classification task are conducted (a) the impact of resizing, padding, and color conversion during the preprocessing step; (b) the effectiveness of transfer learning and fine-tuning in case of limitation of resources and imbalanced classes; (c) the influence of image quality, image quantity, and class labeling accuracy. Three collections of historical palm leaf manuscripts i.e old Sundanese, old Balinese, and old Khmer were investigated. For transfer learning two handwriting datasets i.e MNIST and Omniglot were included. We propose a robust pipeline for handling old scripts by leveraging an ensemble of customized CNN models to improve isolated glyph recognition. While both color padding and resizing method showed minimal performance variation, adjusting the image size proved crucial for achieving optimal results. Additionally, selecting the appropriate pre-trained model, fine-tuning it, and applying soft voting enabled our approach to consistently outperform the baseline across all test scenarios. As a result, the isolated glyph recognition performance on the Old Sundanese, Old Balinese and Old Khmer datasets reached 97.57%, 94.96%, and 92.19%, respectively.

Deep survival models are increasingly used for time-to-event prediction under censoring, yet training objectives remain a bottleneck. The widely used discrete-time negative log-likelihood (NLL) supervises hazards and can suffer from temporal information imbalance and gradient attenuation, yielding early-dominated probability mass and degraded late-horizon calibration, especially under heavy censoring and competing risks. We introduce Pathwise Cross-Entropy (PCE), which utilizes a symmetric, full-path objective that directly learns the occurred-by-t trajectory as a Cumulative Incidence Function (CIF). This direct approach seamlessly yields a normalized Probability Mass Function (PMF) for predicting event times, unlike NLL, where the derived PMF is structurally biased toward monotonic decrease, hindering its predictive utility. In a counting-process view, PCE supplies bidirectional gradients and constitutes a strictly proper scoring rule on counting paths. We extend PCE to competing risks with cause-specific supervision that avoids the multinomial coupling in NLL under competing risks. Empirically, across the tabular SEER and a WSI-derived kidney dataset and multiple backbones, PCE consistently improves discrimination (C-index, AUC) and calibration (IBS), produces calibration plots (ECE and PP plots) that are closer to observation, and enables ordinal first-hit time prediction directly with minimal practical monotonicity violations. These results indicate that PCE is a reliable and interpretable objective for single and competing-risk survival.

Interactive Play Spaces can support positive behaviour. Play endings, user experience (UX), and replay intention, can play an important role to achieve this. However, the relation between these aspects is underexplored. We explore how different types of endings–open, closed positive (winning), and closed negative (losing)–affect user experience and attitude towards replay in a semi-situated study with 93 adults in a science center. While assigned ending conditions did not significantly influence reported experience, many participants, significantly in the open-ended condition, perceived their assigned ending condition differently. Analysis on these self-reported endings revealed that players who experienced a closed negative ending reported higher Stimulation (UEQ). Additionally, user experience dimensions (Attractiveness, Dependability, Stimulation) and Positive Affect (I-PANAS-SF) positively related to attitude towards replay. These findings provide insights into the relation between play endings, UX and attitude towards replay, and highlight the importance of inquiring about experienced experimental conditions in user research.

The Participant Journey Map (PJM) provides structured insight into participation with interactive play in (semi-) public environments. It supports understanding of participants’ behavior and was developed based on experiences with previously developed playful interfaces, related research and expert interviews. We apply the PJM to interactive playful museum exhibits and evaluate and refine it based on its usage in a situated context. We observed 672 play sessions with 6 interactive playful museum exhibits. The observation data was visualized and analyzed using the PJM. This study shows that the PJM provides a realistic representation of participant behaviour, can be used to identify stagnations and progressions in participation flow, and support identification of influencing design and contextual factors. With this paper we contribute by presenting the PJM as a well-grounded, valuable and realistic framework for evaluating and understanding participation with situated interactive play, based on post-hoc evaluation of multiple interfaces with many users.

Pollen classification is considered an important task in palynology. In the Netherlands, two genera of the Urticaceae family, named Parietaria and Urtica, have high morphological similarities but induce allergy at a very different level. Therefore, distinction between these two genera is very important. Within this group, the pollen of Urtica membranacea is the only species that can be recognized easily under the microscope. For the research presented in this study, we built a dataset from 6472 pollen images and our aim was to find the best possible classifier on this dataset by analysing different classification methods, both machine learning and deep learning-based methods. For machine learning-based methods, we measured both texture and moment features based on images from the pollen grains. Varied feature selection techniques, classifiers as well as a hierarchical strategy were implemented for pollen classification. For deep learning-based methods, we compared the performance of six popular Convolutional Neural Networks: AlexNet, VGG16, VGG19, MobileNet V1, MobileNet V2 and ResNet50. Results show that compared with flat classification models, a hierarchical strategy yielded the highest accuracy with 94.5% among machine learning-based methods. Among deep learning-based methods, ResNet50 achieved an accuracy of 99.4%, slightly outperforming the other neural networks investigated. In addition, we investigated the influence on performance by changing the size of image datasets to 1000 and 500 images, respectively. Results demonstrated that on smaller datasets, ResNet50 still achieved the best classification performance. An ablation study was implemented to help understanding why the deep learning-based methods outperformed the other models investigated. Using Urticaceae pollen as an example, our research provides a strategy of selecting a classification model for pollen datasets with highly similar pollen grains to support palynologists and could potentially be applied to other image classification tasks.

Text line segmentation is an important step in the historical document image analysis pipeline to supply useful information for recognition, keyword spotting and indexing. Many handcrafted-based and learning-based approaches have been developed to cope with text line extraction challenges. In this work we present a hybrid technique which combines a convolutional-based denoising task and heuristic Seam Carving framework. We propose the following changes to the original Seam Carving: (1) We applied adaptive slice to anticipate miss-extraction on a short text during medial seam computation. (2) We applied a triple smoothing to find the best local maxima of the smoothed horizontal projection profile which represents candidate medial seams. (3) We utilized five post-processing steps aimed at reconstructing a more precise medial seam. Three different palm leaf data sets: old Sundanese, old Balinese, and old Khmer, have been used to compare Convolutional Seam Carving (CSC) to several baseline methods, including the original Seam Carving. Experimental results show that the proposed method outperforms other current handcrafted-based baselines on all three palm leaf manuscript (PLM) data sets. On the old Sundanese data sets, CSC can produce a significant improvement of the performance rate compared to all other baseline approaches, and it also enhance the measurement on old Balinese and old Khmer datasets. In the ablation study, we discovered that a foreground extraction step is not only able to reduce noises and color degradation but also provide better separation of text region. Following that, an adaptive slice and triple smoothing approach contribute to solve the text length variation problem. Finally, post-processing steps were effective in connecting discontinuous medial seam. The code has been published in https://github.com/erickpaulus/text-line-segmentation.

Automated nanoparticle phenotyping is a critical aspect of high-throughput drug research, which requires analyzing nanoparticle size, shape, and surface topography from microscopy images. To automate this process, we present an instance segmentation pipeline that partitions individual nanoparticles on microscopy images. Our pipeline makes two key contributions. Firstly, we synthesize diverse and approximately realistic nanoparticle images to improve robust learning. Secondly, we improve the BlendMask model to segment tiny, overlapping, or sparse particle images. Specifically, we propose a parameterized approach for generating novel pairs of single particles and their masks, encouraging greater diversity in the training data. To synthesize more realistic particle images, we explore three particle placement rules and an image selection criterion. The improved one-stage instance segmentation network extracts distinctive features of nanoparticles and their context at both local and global levels, which addresses the data challenges associated with tiny, overlapping, or sparse nanoparticles. Extensive experiments demonstrate the effectiveness of our pipeline for automating nanoparticle partitioning and phenotyping in drug research using microscopy images.

Neutrophils are one of the Neutrophils are one of the crucial immune cells. It plays a key role in the immune system defending the invasion of harmful particles, such as viruses and bacteria. The analysis of the dynamic process of neutrophil migration helps biologists to understand underlying mechanisms of neutrophils in response to wounding. However, accurate neutrophil tracking is still a challenging task in 3D space due to the complexity of cell morphology and behavior. In this study, we improved the quality of raw data by denoising and linear interpolation. A 3D U-Net was trained and used to detect the location of each cell in the time-lapse movie. Subsequently, a feature weighted 3D tracking method was proposed. The experimental results show that our method performs well compared to the existing tracking algorithm. Our pipeline is also much more reproducible than other state-of-arts.

As of 2021, more demanding NOx emission requirements entered into force for newly built ships operating on the North and Baltic Sea. Even though various methods are used to assess ships' pollution in ports and off the coastal areas, monitoring over the open sea has been infeasible until now. In this work, we present a novel automated method for evaluation of NO2 emissions produced by individual seagoing ships. We use the spatial association statistic local Moran's I in order to improve the distinguishability between the plume and the background. Using the Automatic Identification Signal (AIS) data of ship locations as well as incorporated uncertainties in wind speed and wind direction, we automatically associate the detected plumes with individual ships. We evaluate the quality of ship-plume matching by calculating the Pearson correlation coefficient between the values of a model-based emission proxy and the estimated NO2 concentrations. For five of the six analyzed areas, our method yields results that are an improvement over the baseline approach used in a previous study.

MOTIVATION: The Flavivirus genus includes several important pathogens, such as Zika, dengue and yellow fever virus. Flavivirus RNA genomes contain a number of functionally important structures in their 3' untranslated regions (3'UTRs). Due to the diversity of sequences and topologies of these structures, their identification is often difficult. In contrast, predictions of such structures are important for understanding of flavivirus replication cycles and development of antiviral strategies. RESULTS: We have developed an algorithm for structured pattern search in RNA sequences, including secondary structures, pseudoknots and triple base interactions. Using the data on known conserved flavivirus 3'UTR structures, we constructed structural descriptors which covered the diversity of patterns in these motifs. The descriptors and the search algorithm were used for the construction of a database of flavivirus 3'UTR structures. Validating this approach, we identified a number of domains matching a general pattern of exoribonuclease Xrn1-resistant RNAs in the growing group of insect-specific flaviviruses. AVAILABILITY AND IMPLEMENTATION: The Leiden Flavivirus RNA Structure Database is available at https://rna.liacs.nl. The search algorithm is available at https://github.com/LeidenRNA/SRHS. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

BACKGROUND: The hallmarks of cancer provide a highly cited and well-used conceptual framework for describing the processes involved in cancer cell development and tumourigenesis. However, methods for translating these high-level concepts into data-level associations between hallmarks and genes (for high throughput analysis), vary widely between studies. The examination of different strategies to associate and map cancer hallmarks reveals significant differences, but also consensus. RESULTS: Here we present the results of a comparative analysis of cancer hallmark mapping strategies, based on Gene Ontology and biological pathway annotation, from different studies. By analysing the semantic similarity between annotations, and the resulting gene set overlap, we identify emerging consensus knowledge. In addition, we analyse the differences between hallmark and gene set associations using Weighted Gene Co-expression Network Analysis and enrichment analysis. CONCLUSIONS: Reaching a community-wide consensus on how to identify cancer hallmark activity from research data would enable more systematic data integration and comparison between studies. These results highlight the current state of the consensus and offer a starting point for further convergence. In addition, we show how a lack of consensus can lead to large differences in the biological interpretation of downstream analyses and discuss the challenges of annotating changing and accumulating biological data, using intermediate knowledge resources that are also changing over time.

BACKGROUND: Cardiotoxicity, characterized by severe cardiac dysfunction, is a major problem in patients treated with different classes of anticancer drugs. Development of predictable human-based models and assays for drug screening are crucial for preventing potential drug-induced adverse effects. Current animal in vivo models and cell lines are not always adequate to represent human biology. Alternatively, human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) show great potential for disease modelling and drug-induced toxicity screenings. Fully automated high-throughput screening of drug toxicity on hiPSC-CMs by fluorescence image analysis is, however, very challenging, due to clustered cell growth patterns and strong intracellular and intercellular variation in the expression of fluorescent markers. RESULTS: In this paper, we report on the development of a fully automated image analysis system for quantification of cardiotoxic phenotypes from hiPSC-CMs that are treated with various concentrations of anticancer drugs doxorubicin or crizotinib. This high-throughput system relies on single-cell segmentation by nuclear signal extraction, fuzzy C-mean clustering of cardiac α-actinin signal, and finally nuclear signal propagation. When compared to manual segmentation, it generates precision and recall scores of 0.81 and 0.93, respectively. CONCLUSIONS: Our results show that our fully automated image analysis system can reliably segment cardiomyocytes even with heterogeneous α-actinin signals.

Biology is 3D. Therefore, it is important to be able to analyze phenomena in a spatiotemporal manner. Different fields in computational sciences are useful for analysis in biology; i.e. image analysis, pattern recognition and machine learning. To fit an empirical model to a higher abstraction, however, theoretical computer science methods are probed. We explore the construction of empirical 3D graphical models and develop abstractions from these models in L-systems. These systems are provided with a profound formalization in a grammar allowing generalization and exploration of mathematical structures in topologies. The connections between these computational approaches are illustrated by a case study of the development of the lactiferous duct in mice and the phenotypical effects from different environmental conditions we can observe on it. We have constructed a workflow to get 3D models from different experimental conditions and use these models to extract features. Our aim is to construct an abstraction of these 3D models to an L-system from features that we have measured. From our measurements we can make the productions for an L-system. In this manner we can formalize the arborization of the lactiferous duct under different environmental conditions and capture different observations. All considered, this paper illustrates the joint of empirical with theoretical computational sciences and the augmentation of the interpretation of the results. At the same time, it shows a method to analyze complex 3D topologies and produces archetypes for developmental configurations.

Large collections of historical biodiversity expeditions are housed in natural history museums throughout the world. Potentially they can serve as rich sources of data for cultural historical and biodiversity research. However, they exist as only partially catalogued specimen repositories and images of unstructured, non-standardised, hand-written text and drawings. Although many archival collections have been digitised, disclosing their content is challenging. They refer to historical place names and outdated taxonomic classifications and are written in multiple languages. Efforts to transcribe the hand-written text can make the content accessible, but semantically describing and interlinking the content would further facilitate research. We propose a semantic model that serves to structure the named entities in natural history archival collections. In addition, we present an approach for the semantic annotation of these collections whilst documenting their provenance. This approach serves as an initial step for an adaptive learning approach for semi-automated extraction of named entities from natural history archival collections. The applicability of the semantic model and the annotation approach is demonstrated using image scans from a collection of 8, 000 field book pages gathered by the Committee for Natural History of the Netherlands Indies between 1820 and 1850, and evaluated together with domain experts from the field of natural and cultural history.

Accurate segmentation of zebrafish from bright-field microscope images is crucial to many applications in the life sciences. Early zebrafish stages are used, and in these stages the zebrafish is partially transparent. This transparency leads to edge ambiguity as is typically seen in the larval stages. Therefore, segmentation of zebrafish objects from images is a challenging task in computational bio-imaging. Popular computational methods fail to segment the relevant edges, which subsequently results in inaccurate measurements and evaluations. Here we present a hybrid method to accomplish accurate and efficient segmentation of zebrafish specimens from bright-field microscope images. We employ the mean shift algorithm to augment the colour representation in the images. This improves the discrimination of the specimen to the background and provides a segmentation candidate retaining the overall shape of the zebrafish. A distance-regularised level set function is initialised from this segmentation candidate and fed to an improved level set method, such that we can obtain another segmentation candidate which preserves the explicit contour of the object. The two candidates are fused using heuristics, and the hybrid result is refined to represent the contour of the zebrafish specimen. We have applied the proposed method on two typical datasets. From experiments, we conclude that the proposed hybrid method improves both efficiency and accuracy of the segmentation of the zebrafish specimen. The results are going to be used for high-throughput applications with zebrafish.

In life sciences, light microscopy is used to study specimens. On the organism-level a bright-field representation present an overview for the whole shape of a specimen; the organ-level fluorescent staining representation supports in the interpretation of the detailed intrinsic structures. We present light microscopy axial-view imaging based on the Vertebrate Automated Screening Technology to acquire axial-view images for the organism and organs of zebrafish larvae. We obtain multi-modal 3D reconstruction using a profile-based method, from which we can derive the 3D measurements of volume and surface area. In this method, we employ a microscope camera calibration using voxel residual volume maximization algorithm. We intuitively align and fuse the obtained multi-models. Experimental results show natural visualization both for the whole organism and organ of zebrafish larvae; subsequently accurate 3D measurements are obtained. This method is very suitable for high-throughput research in which knowledge on size and shape is relevant to the understanding for development, effects of compounds or drugs.

In the early stages of 3D design, sketches are used to quickly conceptualize ideas and gain insight into problems and possible solutions. Computer-aided design tools are widely used for 3D modeling and design, but their required precision and 2D mouse and screen-based interface inhibit the flow of ideas. A study was conducted to explore the efficiency of hand tracking and virtual reality (VR) for 3D object manipulations in conceptual design. Based on existing research on conceptual design and hand gestures, an intuitive hand-based interaction model is proposed. An experiment on basic 3D manipulation shows that participants using a simple VR and hand-tracking interface prototype have similar performance to those using a traditional mouse and screen interface. For the improvement of gestural conceptual design interfaces, the relevant issues are identified.

In order to preserve sufficient fluorescence intensity and improve the quality of fluorescence images in optical projection tomography (OPT) imaging, a feasible acquisition solution is to temporally formalize the fluorescence and bright-field imaging procedure as two consecutive phases. To be specific, fluorescence images are acquired first, in a full axial-view revolution, followed by the bright-field images. Due to the mechanical drift, this approach, however, may suffer from a deviation of center of rotation (COR) for the two imaging phases, resulting in irregular 3D image fusion, with which gene or protein activity may be located inaccurately. In this paper, we address this problem and consider it into a framework based on sinogram unification so as to precisely fuse 3D images from different channels for CORs between channels that are not coincident or if COR is not in the center of sinogram. The former case corresponds to the COR deviation above; while the latter one correlates with COR alignment, without which artefacts will be introduced in the reconstructed results. After sinogram unification, inverse radon transform can be implemented on each channel to reconstruct the 3D image. The fusion results are acquired by mapping the 3D images from different channels into a common space. Experimental results indicate that the proposed framework gains excellent performance in 3D image fusion from different channels. For the COR alignment, a new automated method based on interest point detection and included in sinogram unification, is presented. It outperforms traditional COR alignment approaches in combination of effectiveness and computational complexity.

BACKGROUND: Endocytosis is regarded as a mechanism of attenuating the epidermal growth factor receptor (EGFR) signaling and of receptor degradation. There is increasing evidence becoming available showing that breast cancer progression is associated with a defect in EGFR endocytosis. In order to find related Ribonucleic acid (RNA) regulators in this process, high-throughput imaging with fluorescent markers is used to visualize the complex EGFR endocytosis process. Subsequently a dedicated automatic image and data analysis system is developed and applied to extract the phenotype measurement and distinguish different developmental episodes from a huge amount of images acquired through high-throughput imaging. For the image analysis, a phenotype measurement quantifies the important image information into distinct features or measurements. Therefore, the manner in which prominent measurements are chosen to represent the dynamics of the EGFR process becomes a crucial step for the identification of the phenotype. In the subsequent data analysis, classification is used to categorize each observation by making use of all prominent measurements obtained from image analysis. Therefore, a better construction for a classification strategy will support to raise the performance level in our image and data analysis system. RESULTS: In this paper, we illustrate an integrated analysis method for EGFR signalling through image analysis of microscopy images. Sophisticated wavelet-based texture measurements are used to obtain a good description of the characteristic stages in the EGFR signalling. A hierarchical classification strategy is designed to improve the recognition of phenotypic episodes of EGFR during endocytosis. Different strategies for normalization, feature selection and classification are evaluated. CONCLUSIONS: The results of performance assessment clearly demonstrate that our hierarchical classification scheme combined with a selected set of features provides a notable improvement in the temporal analysis of EGFR endocytosis. Moreover, it is shown that the addition of the wavelet-based texture features contributes to this improvement. Our workflow can be applied to drug discovery to analyze defected EGFR endocytosis processes.

Both Petri nets and differential equations are important modeling tools for biological processes. In this paper we demonstrate how these two modeling techniques can be combined to describe biological gradient formation. Parameters derived from partial differential equation describing the process of gradient formation are incorporated in an abstract Petri net model. The quantitative aspects of the resulting model are validated through a case study of gradient formation in the fruit fly.

Digital brain atlases are used in neuroscience to characterize the spatial organization of neuronal structures [1]–[3], for planning and guidance during neurosurgery [4], [5], and as a reference for interpreting other modalities such as gene expression or proteomic data [6]–[9]. The field of digital atlasing is extensive, and includes high quality brain atlases of the mouse [10], rat [11], rhesus macaque [12], human [13], [14], and several other model organisms. In addition to atlases based on histology, [11], [15], [16], magnetic resonance imaging [10], [17], and positron emission tomography [11], modern digital atlases often use probabilistic and multimodal techniques [18], [19], as well as sophisticated visualization software [20], [21]. Whether atlases involve detailed visualization of structures of a single or small group of specimens [6], [22], [23] or averages over larger populations [18], [24], much of the work in developing digital brain atlases is from the perspective of the user of a single resource. This is often due largely to the challenges of data generation, maintenance, and resources management [25], [26]. A more recent goal of many neuroscientists is to connect multiple and diverse resources to work in a collaborative manner using an atlas based framework [2], [19]. This vision is appealing as, ideally, researchers would be able to share their data and analyses with others, regardless of where they or the data are located. An important step in this direction is the specification of a common frame of reference across specimens and resources (either as coordinate, ontology, or region of interest) that is adopted by the community. In this perspective, we propose a collaborative digital atlasing framework for coordinating mouse brain research that allows access to data, tools, and analyses from multiple sources.

miRNAs are small regulators that mediate gene expression and each miRNA regulates specific target genes. In animals, target prediction of the miRNAs is accomplished through several computational methods, i.e. miRanda, TargetScan and PicTar. Typically, these methods predict targets from features of miRNA-target interaction such as sequence complementarity, free energy of RNA duplexes and conservation of target sites. They are constructed for high throughput and also result in a large amount of predictions and a high estimated false-positive rate. To date, specific rules to capture all known miRNA targets have not been devised. We observed that miRNAs sometimes share targets. Therefore, in this paper we present an approach which analyzes miRNA-miRNA relationships and utilizes them for target prediction.We use machine learning techniques to reveal the feature patterns between known miRNAs. Different data setups are evaluated and compared to achieve the best performance. Furthermore, the derived rules are applied to miRNAs of which the targets are not yet known so as to see if new targets could be predicted. In the analysis of functionally similar miRNAs, we found that genomic distance and seed similarity between miRNAs are dominant features in the description of a group of miRNAs binding the same target. Application of one specific rule resulted in the prediction of targets for seven miRNAs for which the targets were formerly unknown. Some of these targets were also detected by the existing methods. Our method contributes to the improvement of target identification by predicting targets with high specificity and without conservation limitation.

The paper discusses the use of XML as an intermediate format for storage of information on microscope images and the different techniques to visualize this information. In this manner we have explored XML and built several applications for image management. The XML-files generated by these applications are defined as the lab-notebook. A crucial feature of such a lab-notebook stored in XML is that it directly relates to the Internet so that a modern Web browser can be used to present the data in a user-friendly manner. We discuss how we used XSL and Java applets to accomplish this.

Utilizes the Euclidean distance transform as a tool for shape analysis and description. The author shows effective use of some (new) properties of the vector distance transform in determining parameters for a two-step transform of a distorted image to its undistorted equivalent. The transform is based on the outer contour of the object of interest as present in the image.>