Support for Amazon Textract

While the source data from client systems is typically available in the form of data files or extracts, there are some instances in which program data related to children and families is available in the form of images or PDFs of either electronically filled or handwritten forms. Extracting this type of data, particularly handwritten data, from PDKs is often a labor-intensive and error-prone process, especially when it is done manually. CUSP now offers support for the Amazon Textract tool that detects typed an handwritten text from a variety of document formats such as images and PDF files. The tool extracts text, forms, and tables from such documents using the Amazon Textract Document Analysis API.

  • CUSP now supports the use of Amazon Textract to automate the extraction of handwritten content from PDF documents and image files, dramatically reducing the time and effort required for data entry for ingestion into CUSP while also enhancing the accuracy of data entry.
A visual representation of a handwritten document's data going into CUSP.


  • With Amazon Textract, CUSP now automates the extraction of handwritten data from PDF and image files, thus drastically reducing the time and effort required for manual data entry. The result is faster
    data availability and more seamless data ingestion from a variety of data sources and formats.
  •  CUSP leverages handwriting recognition algorithms in Amazon Textract to adapt and learn from patterns in handwriting. This ensures a higher level of accuracy in deciphering handwritten content
    compared to what is generally achievable with human transcribers.
  • The use of Amazon Textract for data extraction and ingestion in CUSP offers scalability by allowing the processing of a large number of forms in a short span of time without compromising accuracy. This
    ensures consistent and reliable data extraction.