Küper A, Lodde G, Livingstone E, Schadendorf D, Krämer N. Mitigating cognitive bias with clinical decision support systems: an experimental study. Journal of Decision Systems 2023. 1–20. https://doi.org/10.1080/12460125.2023.2245215
Küper A, Krämer N. Psychological Traits and Appropriate Reliance: Factors Shaping Trust in AI. International Journal of Human–Computer Interaction 2024. 1–17. https://doi.org/10.1080/10447318.2024.2348216
Küper A, Lodde G, Livingstone, E, Schadendorf D, Krämer N. Psychological Traits Influencing Appropriate Reliance on AI-enabled Clinical Decision Support Systems: an Experimental Study. Journal of Medical Internet Research 2024. submitted
FHIR-PYrate
FHIR-PYrate is a Python package designed to query FHIR servers and return structured data as pandas DataFrames. It supports filtering resources using regular expressions and SpaCy, and can download DICOM studies and series. The package is particularly useful for healthcare data scientists needing to manipulate and analyze FHIR data efficiently.
DeidentiFHIR
DeidentiFHIR is a Scala/Java tool designed for de-identifying FHIR resources to ensure privacy in healthcare data. It provides customizable de-identification methods while maintaining the usability of the data, such as masking personal identifiers or replacing them with pseudonyms. This tool is essential for securely handling patient data in compliance with privacy regulations.
DeidentiFHIR-Pipeline
With the DeidentiFHIR-Pipeline, you can transfer FHIR based data from one source (e.g. a FHIR server) to a target (e.g. a FHIR server) and pseudonymize the data in between. Pseudonymization is based on the DeidentiFHIR library.
The transfer consists of four steps:
There can be multiple implementations for each step, e.g. a cohort selection could be based on consent policies stored in gICS or based on a list of IDs (e.g. Patient Identifiers).
DicomDeidentify
DicomDeidentify is a tool and library written in Scala designed for anonymizing DICOM files while preserving the usability of the medical imaging data. It offers customizable de-identification processes, enabling users to remove or modify personal identifiers to comply with privacy standards. This tool is essential for secure handling and sharing of medical images in both research and clinical environments.
dicom-deidentifier-ts
The dicom-deidentifier-ts is a TypeScript-based tool designed for anonymizing DICOM files, focusing on removing or obfuscating personal health information to comply with privacy regulations. It provides a robust framework for configuring and executing de-identification processes on medical imaging data. This tool is essential for securely managing and sharing DICOM files in healthcare environments.
Datavzrd
Datavzrd is a zero/low-code, interactive, visual, server-free, browser-based reporting tool for tabular datasets. It allows users to create visually enriched, standalone HTML reports with various features like heatmaps, tick plots, and external resource links, all configurable via YAML. Datavzrd enables scalable, shareable, and reproducible data exploration without the need for specialized web applications or server maintenance.
FHIR-Converter
FHIR-Converter is a FHIR Data Conversion Web Service designed to implement the $convert-data custom FHIR (Fast Healthcare Interoperability Resources) operation, enabling the conversion of HL7v2, CCDA, JSON, and FHIR STU3 into FHIR R4 format. This service is built using the Microsoft FHIR Converter and is called similarly to the custom $convert-data FHIR operation described in Microsoft FHIR Server documentation. You send a request to our FHIR server specifying the data conversion you want to perform. The server then converts your input data into the FHIR R4 format, making it compatible with systems and applications that support this standard.
doctext
DocText is a Python package that extracts and processes text from various document formats, such as PDFs, Word files, and images. It provides functionalities for optical character recognition (OCR) and text extraction, making it useful for automating document analysis tasks. This tool is essential for developers working on projects involving text extraction and processing from diverse document types.
HUFF
HUFF (Human-friendly FHIR) is a tool primarily written in Rust that converts complex FHIR JSON data structures into a more readable YAML format, making it easier to understand and navigate the data. It simplifies the dense and nested JSON trees by condensing repetitive patterns and converting them into a more intuitive structure. The tool is not a replacement for JSON but offers an alternative way to view and work with FHIR data, enhancing its accessibility for users.
orthanq Orthanq (ORThogonal evidence based HAplotype Quantification) is a tool for uncertainy-aware haplotype quantification at subclone resolution. It can be used to perform HLA typing of normal (healthy) or tumor cells as well as virus lineage quantification. Orthanq is available on Bioconda and it can be reached at: https://orthanq.github.io/.
orthanq-evaluation An evaluation workflow that performs HLA typing using Orthanq. HLA typing is achieved for given or simulated samples based on the user configuration in the config. The three scenarios that are considered in the workflow include simulation of samples given HLA alleles in alleles sheet, generation of subclonal samples with samples given in sample sheet and haplotype quantification based on given samples in sample sheet.
CLAT-cross-lingual-annotation-transfer Cross Language Transfer - Translating data records: Translating datasets with their annotations between multiple target languages. DOI: https://doi.org/10.18653/v1/2022.clinicalnlp-1.6
BioKGrapher BioKGrapher is a comprehensive tool designed for the automatic construction of knowledge graphs (KGs) from large-scale biomedical literature, processing PubMed IDs as input.
Coming soon