MedicalDataHandler, a research-oriented graphical user interface for DICOM data management.

Intracranial tumors are the most common neoplasms of childhood, accounting for approximately 20% of all pediatric malignancies. Radiation therapy has led directly to significant increases in survival of children with certain types of intracranial tumors; however, given the aggressive nature of this therapy, children are at risk for exhibiting changes in brain structure, neuronal biochemistry, and neurocognitive functioning. In this case report, we present neuropsychological, magnetic resonance imaging, proton magnetic resonance spectroscopic imaging, and diffusion tensor imaging data for two adolescents (one patient with ependymal spinal cord tumor with intracranial metastases, and one healthy, typically developing control) from three time points as defined by the patient's radiation schedule (baseline before the patient's radiation therapy, 6 months following completion of the patient's radiation, and 27 months following the patient's radiation). In the patient, there were progressive decreases in gray and white matter volumes as well as early decreases in mean N-acetyl aspartate/choline (NAA/Cho) ratios and fractional anisotropy (FA) in regions with normal appearance on conventional MRI. At the last follow-up, NAA/Cho and FA tended to change in the direction to normal values in selected regions. At the same time, the patient had initial reduction in language and motor skills, followed by return to baseline, but later onset delay in visuospatial and visual perceptual skills. Results are discussed in terms of sensitivity of the four techniques to early and late effects of treatment, and avenues for future investigations.

The Development and Implementation of Radiation Oncology IQ Script Enabled Plans at the Odette Cancer Centre

Background and objective: Despite fast evolution cycles in deep learning methodologies for medical imaging in radiotherapy, auto-segmentation solutions rarely run in clinics due to the lack of open-source frameworks feasible for processing DICOM RT Structure Sets. Besides this shortage, available open-source DICOM RT Structure Set converters rely exclusively on 2D reconstruction approaches leading to pixelated contours with potentially low acceptance by healthcare professionals. PyRaDiSe, an open-source, deep learning framework independent Python package, addresses these issues by providing a framework for building auto-segmentation solutions feasible to operate directly on DICOM data. In addition, PyRaDiSe provides profound DICOM RT Structure Set conversion and processing capabilities; thus, it applies also to auto-segmentation-related tasks, such as dataset construction for deep learning model training.Methods: The PyRaDiSe package follows a holistic approach and provides DICOM data handling, deep learning model inference, pre-processing, and post-processing functionalities. The DICOM data handling allows for highly automated and flexible handling of DICOM image series, DICOM RT Structure Sets, and DICOM registrations, including 2D-based and 3D-based conversion from and to DICOM RT Structure Sets. For deep learning model inference, extending given skeleton classes is straightforwardly achieved, allowing for employing any deep learning framework. Furthermore, a profound set of pre-processing and post-processing routines is included that incorporate partial invertibility for restoring spatial properties, such as image origin or orientation.Results: The PyRaDiSe package, characterized by its flexibility and automated routines, allows for fast deployment and prototyping, reducing efforts for auto-segmentation pipeline implementation. Furthermore, while deep learning model inference is independent of the deep learning framework, it can easily be integrated into famous deep learning frameworks such as PyTorch or Tensorflow. The developed package has successfully demonstrated its capabilities in a research project at our institution for organs-at-risk segmentation in brain tumor patients. Furthermore, PyRaDiSe has shown its conversion performance for dataset construction.Conclusions: The PyRaDiSe package closes the gap between data science and clinical radiotherapy by enabling deep learning segmentation models to be easily transferred into clinical research practice. PyRaDiSe is available on https://github.com/ubern-mia/pyradise and can be installed directly from the Python Package Index using pip install pyradise.

Evaluation of effectives doses to patients after whole body computed tomography

The organisation, conversion, cleaning and processing of DICOM data is an ongoing challenge across medical image analysis research projects. PyDicer (PYthon Dicom Image ConvertER) was created as a generalisable tool for use across a variety of radiotherapy research projects. This includes the conversion of DICOM objects into a standardised form as well as functionality to visualise, clean and analyse the converted data. The generalisability of PyDicer has been demonstrated by its use across a range of projects including the analysis of radiotherapy dose metrics and radiomics features as well as auto-segmentation training, inference and validation.

Automatic text summarization is increasingly explored to manage the large influx of resumes in recruitment, potentially aiding downstream tasks like job category classification. While quantitative evaluations offer insights into summarizer performance, they often do not fully explain why certain methods excel or falter in preserving category-distinguishing information. This paper presents an in-depth qualitative analysis of summaries generated by representative extractive (Reduction, Random) and abstractive (BART, PEGASUS) techniques for resume analysis. We meticulously examine how well crucial information such as specific skills, job titles, and industry-related terms is retained or lost in the generated summaries, and how these characteristics correlate with the performance of a downstream job category classifier. Through illustrative examples and analysis of misclassification patterns, our findings reveal distinct strategies of information preservation and common pitfalls, offering a nuanced understanding beyond aggregate metrics. This qualitative perspective underscores the importance of information fidelity for specific downstream applications and provides actionable insights for selecting or developing summarization tools for effective resume processing in e-recruitment systems.

Three-photon microscopy (3PM) enables high-resolution three-dimensional (3D) imaging in deeply situated and highly scattering biological specimens, facilitating precise characterization of biological morphology and cellular-level physiology in vivo. However, the use of fluorescent probes with relatively low three-photon absorption cross-sections necessitates high-peak-power lasers for excitation, which poses inherent risks of light-induced damage. Additionally, the low repetition frequency of these lasers prolongs scanning time per pixel, hampering imaging speed and exacerbating the potential for photodamage. Such limitations hinder the application of 3PM in studying vulnerable tissues, including muscle regeneration. To address this critical issue, we developed the Multi-Scale Attention Denoising Network (MSAD-Net), a precise and versatile denoising network suitable for diverse structures and varying noise levels. Our network enables the use of lower excitation power (1/4–1/2 of the common power: 1.0–1.5 mW vs 4–6 mW) and shorter scanning time (1/6–1/4 of the common time: 2–3 μs/pixel vs 12 μs/pixel) in 3PM while preserving image quality and tissue integrity. It achieves a structural similarity index (SSIM) of with an average of 0.9932 and a fast inference time of just 80 ms per frame which ensured both high fidelity and practicality for downstream applications. By utilizing MSAD-Net-assisted imaging, we characterize the biological morphology and functionality of muscle regeneration processes through deep in vivo five-channel imaging under low excitation power and short scanning time, while maintaining a high signal-to-noise ratio (SNR) and excellent axial spatial resolution. Furthermore, we conducted high axial-resolution dynamic imaging of vascular microcirculation, macrophages, and ghost fibers. Our findings provide a deeper understanding of the mechanisms underlying muscle regeneration at the cellular and tissue levels.

Partial-arch intraoral scans (IOSs) are commonly used in clinical dentistry where high precision and reduced scanning areas are required. However, most existing tooth segmentation algorithms are developed only for full-arch IOSs and perform poorly when applied to partial-arch data. This study aimed to develop a fully automated deep learning (DL) model for tooth segmentation and labeling on both full- and partial-arch IOSs. We collected 600 IOSs (300 full-arch and 300 partial-arch) from a dental clinic. The proposed model was based on a two-stage DL model (ToothInstanceNet), and incorporated four enhancements: (1) artificial partial-arch IOSs, (2) DL-based alignment module, (3) FDI-aware postprocessing algorithm, and (4) real partial-arch IOSs. Model performance was evaluated via 5-fold cross-validation using F1-score, tooth Dice, tooth macro-F1, and macro-IoU. In addition, the model was evaluated with the public Teeth3DS dataset, and we analysed correlations between dental conditions and model errors. The model achieved an F1-score of 0.9908 and 0.9884; tooth Dice of 0.9819 and 0.9862; tooth macro-F1 of 0.9940 and 0.9786; and macro-IoU of 0.9403 and 0.9280 on full- and partial-arch IOSs, respectively. The model also demonstrated superior performance (score = 0.9870) in 3DTeethSeg challenge. Correlation analyses revealed that certain dental conditions, particularly residual roots, residual crowns, missing teeth, and partially erupted teeth, were significantly and positively associated with the model's errors. The current study proposes the first fully automated method for tooth segmentation and FDI labeling on both full- and partial-arch IOSs. The final model demonstrated high accuracy for both scan types, indicating its potential for integration into clinical dental work. This work could aid clinicians in the first step of tooth identification in digital dental workflows, and lays the groundwork for extending the automation of the downstream applications, such as diagnosis and monitoring on partial-arch IOSs.

Getting a grip on how we may age healthily is a central interest of biogerontological research. To this end, a number of academic teams developed platforms for life- and healthspan assessment in Caenorhabditis elegans. These are very appealing for medium- to high throughput screens, but a broader implementation is lacking due to many systems relying on custom scripts for data analysis that others struggle to adopt. Hence, user-friendly recommendations would help to translate raw data into interpretable results. The aim of this communication is to streamline the analysis of data obtained by the WorMotel, an economically and practically appealing screening platform, in order to facilitate the use of this system by interested researchers. We here detail recommendations for the stepwise conversion of raw image data into activity values and explain criteria for assessment of health in C. elegans based on locomotion. Our analysis protocol can easily be adopted by researchers, and all needed scripts and a tutorial are available in S1 and S2 Files.

Reprogrammable Adenosine Deaminase Acting on RNA (ADAR) Sensors (RADARS) control RNA translation in mammalian cells, allowing for noninvasive sensing or perturbation of specific cell types based on transcriptional signatures. Upon base-pairing between a target RNA and a sensor RNA, RADARS leverages ADAR to edit a premature stop codon upstream of a gene of interest, thereby releasing translation of the desired cargo. These design principles enable sequence programmability, allowing RADARS to adapt more easily to new contexts than existing tools for targeting cell types. We describe a detailed protocol for performing experiments with RADARS, including designing, cloning and validating RADARS constructs targeting a transcript of interest. RADARS guide sequences can be designed with an intuitive web interface and cloned into existing constructs for downstream applications including imaging, sorting and sequencing. We outline recommendations for cargo choice, sensor design and ADAR system selection, enabling users to choose the best workflow depending on the desired application. Beginning with sensor design, the selection of top-performing RADARS guides can be completed in ~2 weeks, followed by a desired use case. Convenient engineering and application of RADARS for various applications enable the design and execution of various cell-targeting experiments.

Biological sequence databases are integral to efforts to characterize and understand biological molecules and share biological data. However, when analyzing these data, scientists are often left holding disparate biological currency-molecular identifiers from different databases. For downstream applications that require converting the identifiers themselves, there are many resources available, but analyzing associated loci and variants can be cumbersome if data is not given in a form amenable to particular analyses. Here we present BISQUE, a web server and customizable command-line tool for converting molecular identifiers and their contained loci and variants between different database conventions. BISQUE uses a graph traversal algorithm to generalize the conversion process for residues in the human genome, genes, transcripts and proteins, allowing for conversion across classes of molecules and in all directions through an intuitive web interface and a URL-based web service. BISQUE is freely available via the web using any major web browser (http://bisque.yulab.org/). Source code is available in a public GitHub repository (https://github.com/hyulab/BISQUE). haiyuan.yu@cornell.edu Supplementary data are available at Bioinformatics online.

BackgroundPathogen detection in clinical samples based on 16S metagenomic sequencing technology in microbiology laboratories is an important strategy for clinical diagnosis, public health surveillance, and investigations of outbreaks. However, the implementation of the technology is limited by its accuracy and the time required for bioinformatics analysis. Therefore, a simple, standardized, and rapid analysis pipeline from the receipt of clinical samples to the generation of a test report is needed to increase the use of metagenomic analyses in clinical settings.ResultsWe developed a comprehensive bioinformatics analysis pipeline for the identification of pathogens in clinical samples based on 16S metagenomic sequencing data, named 16SPIP. This pipeline offers two analysis modes (fast and sensitive mode) for the rapid conversion of clinical 16S metagenomic data to test reports for pathogen detection. The pipeline includes tools for data conversion, quality control, merging of paired-end reads, alignment, and pathogen identification. We validated the feasibility and accuracy of the pipeline using a combination of culture and whole-genome shotgun (WGS) metagenomic analyses.Conclusions16SPIP may be effective for the analysis of 16S metagenomic sequencing data for real-time, rapid, and unbiased pathogen detection in clinical samples.

Introduction: R packages like minfi and RnBeads allow efficient processing and high-level statistics of human DNA methylation arrays, wet-lab scientists who wish to verify methylation differences at particular loci or specific gene sets are typically uncomfortable with the command-line. They nevertheless need to visualise, compare and quantify methylation changes against identified genomic features. Therefore, we wished to develop a tool for simple intuitive mapping of methylation differences to UCSC genome browser which allows methylation differences to be quantified across genomic features and multiple datasets. Methods: CandiMeth currently works in two stages: 1) differential methylation outputs from minfi/RnBeads are mapped to UCSC genome browser allowing easy visualisation of methylation changes and integration with external tracks e.g. CGI location or chromatin modifications. 2) Galaxy (www.usegalaxy.org) was used to generate a workflow for interrogating specific genomic intervals for candidate genes using a minimum of steps. Results: CandiMeth allows rapid conversion of differential methylation data into visually intuitive formats which wet- and dry-lab biologists can both assess. Methylation changes which extend beyond sites initially highlighted in the genome-wide analysis can be easily judged, and mapping of changes to different intervals is straightforward. Choosing primers for pyrosequencing is then simple using BLAT. Quantification of changes at chosen regions only requires creation of a one column dataset containing candidate genes, saving time and resources for further downstream analysis. Multiple datasets can be sequentially analysed and outputs are labelled via candidate gene list, sample methylation dataset and gene region to avoid ambiguity. Outputs are also available for use or download through Galaxy, which creates new datasets for each completed task, avoiding destructive edits to original data. Discussion: With increasing use of microarray technologies in methylation analysis, developing a user-friendly way for wet-lab biologists to interpret and integrate array results into their work has become essential. CandiMeth facilitates this by allowing users with basic bioinformatics skills to easily see the location of methylation changes in their dataset and allows them to quantify differences at candidate genes along with easily designing lab verification assays. Many projects in our lab have adapted this approach, establishing it as a non-computationally intensive method of user-friendly bioinformatics.