Plugging the health care knowledge gap - digi-libris

Go to content

Main menu:

Plugging the health care knowledge gap

Where there is no doctor by Dr. David Werner (497 pages, Hesperian rev. 2017) is considered to be the most widely-used health care manual in the world. This classic title is for health workers, clinicians, health educators, midwives, community leaders, and others involved in primary health care delivery and health promotion around the world.

It’s popularity illustrates the global demand for health related information without downplaying the need for experienced health workers who, by the way, also need access to timely information, both evidence based and of practical value.

But a serious shortage of health workers is one of the major constraints to tackling global access to essential health care services is. One of the proposed solutions is task shifting, the process of delegation whereby tasks are moved, where appropriate, to less specialized health workers [WHO 2008]. Specifically health workers should be trained and encouraged to execute certain tasks normally performed by medical professionals, particularly in situations of emergency or when no doctor can be reached.

Enabling health workers to assume higher level tasks and facilitating easy access to know-how may be one of the most promising endeavours to achieve the UN sustainable health goal.

Encouraging health workers to acquire new skills and instilling a sense of purpose is a first step, but the provision of training (professional skills, information literacy, communication technology) as well as tools and support are ongoing responsibilities.

The delivery of actionable knowledge to both experienced and less experienced health workers is an integral part of support, but information overload being a grim reality makes it increasingly difficult for individuals to filter out useful know-how from all available information.
A large body of practical knowledge exists already, but it is often hard to find, deeply buried in a much larger mass of academic content and submerged in advertising, distracting animations and unrelated links.
Health workers depend on rapidly accessing life-saving knowledge (user-friendly textbooks, tutorial videos, course material etc.) in a language they can understand and don’t have the time to wade through hundreds of web pages and guessing where to click to find and download practical advice for off-line use.
This project looks at the three components of the knowledge exchange process (content, exposure/discovery and adoption) and proposes a universally applicable know-how delivery system that is provider neutral, facilitates discovery and comparison before downloading and helps end users to organize their personal knowledge base so that they can rapidly re-find the information they may need in an emergency situation.

Components of knowledge exchange
1.     Content
Useful content in this context is any digital asset available online or offline that a) contains useful learning support or readily implemental practical know-how and b) can classify itself automatically in the user’s knowledge base.
Typical examples are precisely targeted documents and textbooks (usually in PDF format) and tutorial videos (*.mp4 or *.mov), but also course material and telemedicine apps etc.

Content providers offering articles of practical knowledge are encouraged to present their contributions in way to facilitate exposure + discovery and adoption to help closing the health care knowledge gap.

Core to this system are search friendly documents containing catalogue data (aka metadata) that travels with each digital asset, either embedded, in an accompanying data manifest or in a data + metadata bundle zip file - useful in cases where metadata cannot directly be embedded in the digital resource.

It is recommended that keywords are always be listed in English (in addition to the document language) to facilitate searching at the user’s end (assuming their reading software can handle translation).
Since Textbooks with several hundred pages are never read from cover to cover but consulted only in specific situations, it is equally important that such documents are user friendly, i.e. easy to navigate and to peruse. They must absolutely contain bookmarks and an interactive table of contents.

2.     Exposure + discovery
Ideally all items that fall under the category of practical knowledge or know-how are exposed or presented by all content providers in a distinctive way so that they can be instantly located and discovered by health workers on any web site, compared with other offerings and ambiguities eliminated.

This latter point requires that the items include embedded, attached or bundled metadata for indexing and searching purposes at the recipient’s end (at least a meaningful title, description and keywords).

Whether preloaded (boxed) or offered online, free or against payment, as a single contribution or as part of a bulk offering, metadata must accompany every resource offered.
In PDF files these can easily be embedded and viewed in a reader. In office documents and some multimedia files thy can also be embedded, but may require special software to extract.

Where embedding is not possible such metadata can be packaged together with the object in a data+metadata container file for individual items or for bulk submission in a manifest using an electronic library exchange format (see proof of concept below).

3.     Adoption
Adoption follows successful discovery, downloading of an article and integrating it in one’s personal knowledge base for future use. This may require specially enabled software tools (see proof of concept below).

Software developers of knowledge organization tools are encouraged to produce apps for end users that
  • Can Import digital assets of any type and display them in a searchable list by title, topic or other attributes
  • Reads embedded metadata in PDF files as well as in data + metadata bundles and in manifests published for the bulk exchange of library items.
  • Allows the user to add local and downloaded content from any source and adjust its metadata to suit his needs and facilitate searching
Proof of concept
 For the purpose of demonstrating the viability of the proposed project and reaching the objective, we have developed

  • dMeta: for bundling data + metadata in a single package
  • elxf: a universally applicable      electronic library exchange format
  • digi-libris mobile: a free software tool for end users that understands the above formats, can extract metadata and use them to automatically classify incoming material in a user’s knowledge base for easy re-finding when needed in a hurry.
Where embedding is not possible such metadata can be packaged together with the object in a dMeta container file (see for individual items or for bulk submission in a manifest using the electronic library exchange format (see
dMeta stands for Data plus Metadata bundled together in a zip file with the ending *.dMeta.
They are simply assemblies of electronic files of any kind tied together with their corresponding metadata. Such a package may contain multiple pairs of data and metadata sets and even single metadata files containing virtual items or links to an external resource.
dMeta uses the XMP file format for the metadata part, a proven technology developed by Adobe©, same as used inside PDF. XMP files use the XML structure and they can even be built “by hand” with any text processing tool and saved as plain UTF-8 text file.
The format is designed for
  • content providers to distribute any type of digital asset together with its relevant metadata to facilitate discovery and automatic classification in the user’s knowledge base,
  • libraries to distribute the descriptive metadata, which they hold anyhow in their repositories, together with documents which they are not allowed to modify due to checksum verification of the original item,
  • connecting researchers with research in that any file can be coupled and transmitted together with its descriptive metadata. This can include research-relevant data from an instrument and support notes that are not normally included in a paper but that Open Access requires to be made available to the scientific universe. Scientists can more easily share resources and knowledge and it facilitates collaboration,
  • increasing an author’s exposure by making sure that metadata survives peer review and page setup as the publisher can easily import this metadata in the final PDF prior to publishing
elxf  is an electronic library exchange format for personal use and a simple collaborative solution for all content providers (including Open Access publishers and organizations with smaller publishing budgets) and freely accessible by users.

elxf is provider agnostic and enables health professionals to collect all information of interest to them (without having to join dozens of networks), from any speciality field and regardless of who provided it, and display it in a single list for a better overview and to quickly re-find information in the mass of material that accumulates over time. Users control what they want to investigate further (rather than having search engines deciding on their behalf), allowing them to
selectively acquire just the information that is useful to them
  • suppress duplicate, outdated, overlapping and conflicting information
  • reduce application clutter with multiple front ends and proprietary indices
  • update the collection frequently and consult it online and offline (in low bandwidth situations)

elxf is not  about another global repository with a central authority, just an open specification (see that facilitates the exchange of information and allows users to dynamically build up their own searchable knowledge base from downloaded documents and other digital assets as well as links to online resources.
Content providers publish their material on their web site or on removable media along with a manifest (typically named “library.elxf”) containing all metadata pertaining to that material. Users, upon opening the manifest, see instantly more than just cryptic file names (including translated descriptions and possibly a preview) to decide which items are relevant and worth downloading for future reference.

elxf builds on proven standards and technologies such as Dublin Core for parameter naming and Adobes' XMP format for metadata exchange (as already used in PDF files) plus an extensible topics definition to accommodate varying taxonomies.

While the filename itself, the title and the description are the most obvious metadata that have to be meaningful, topics and keywords are equally important for providing better search results and the inclusion of abstracts in several languages or machine translatable at the user’s end can greatly improve the discovery.

digi-libris mobile
digi-libris mobile is a free content viewer for Android devices. It is a non-tech personal organization tool for todays’ multitasking user who works and lives in a digital world. This is metadata centric software for the automatic organization of your personal knowledge base and collection of digital assets of any kind.

Mix text books, tutorial videos, images and other data sets as well as web links to remote resources in a single list and see all vital attributes at a glance to make informed decisions before downloading or opening an item.

It reads embedded metadata in PDF files, in dMeta bundles and in elxf manifests supporting the electronic library exchange format for bulk downloading and automatically classifies added or downloaded items for easy re-finding in the mass of material that accumulates over time.

The user can easily add content from any source, both locally stored and downloaded and edit it’s metadata to suit his needs, e.g. to add keywords to help re-finding an article.

100+ Languages
Information in this context is considered useful when the user a) fully understands in and b) can instantly apply the practical knowledge it contains.
In low-resource settings this can become a real problem, particularly for people with limited mastery of English or language in a given document. This is why digi-libris includes algorithms permitting to

Instantly translate titles, descriptions and abstracts
This makes use of third party (online or offline) services. While machine translations do not pretend to produce literary masterpieces and may not always be absolutely accurate, they nevertheless offer a good idea what the content is all about to let the user decide whether or not to investigate further.
Translate search terms
Since a lot of documentation is available in English only and content providers are encouraged to (also) list keywords in English, a user can enter a query in his own language which will be translated into English prior to launching a search across all items in the database.
Video language prompts
Tutorial videos can save lives and therefore must be easily accessible to all who need them and in a language they understand.

Full text prompts can be added in any language to any video without requiring a new video shooting for each language and they cost a fraction of the cost of other postproduction methods.

Text prompts are list that scroll up in sync with the action and/or the voice-over and videos can be paused for reading. They display on top or on the side of a video and work both for locally stored and streamed movies (incl. YouTube).

The text translations and synchronisation tags are commissioned by the provider and are typically included in a dMeta bundle or added as a linked item in an elxf manifest.
Content providers desiring to offer such help to their audience will advise their readers to search for and download the free digi-libris mobile app on the Google Play Store for use on Android tablets and phones. Desktop versions for Windows, Mac and Linux can be downloaded from

For technical questions contact:
For a license for the Windows Pro version please write to

launching a pilot project in an African county, in India or anywhere else to increase the capacity of health works to acquire new skills and to enable them to intervene when there is no doctor to help.

We are currently looking for content providers (of know-how or tutorial videos) willing to cooperate and  to mount an elxf-based system to deliver life-saving matter faster and more widely to the parties who need it most desperately.

The pilot project will allow us to test the elxf      format in practice and to refine it as needed but also to call for the creation of a steering committee, to show the advantage or even the necessity of having the elxf format curated by a major health NGO and to have it placed on a central server which could facilitate discovery and comparison.
digi-libris Reader is a non-tech personal organization tool for today’s multitasking user who works and lives in a digital world.

This is metadata centric software for the automatic organization of your catalogue of documents and collection of things.

Mix documents and data sets of any type, photos, music, videos and web links in a single list, see vital attributes at a glance. You can also add physical or virtual items manually.

It is free and it is very easy to use: Just drag a file or a web link into the main window and it will automatically be classified.

for more information please contact the digi-libris hotline
Back to content | Back to main menu