We added additional filters to the Explore tool. Now you can filter your search results before starting the search. From the Explore tool page you can choose which datasets you want to search within and restrict the search by the publishing date or repositories.
Fork a map
“Fork a map” allows you to edit a copy of the map by editing the structure of the fingerprint. You can expand all concepts of the map to get a better overview. You can merge and move concepts around to change it as it suits you the most, organize the map and get better control of the content.
Relevant article highlighting
Another useful feature is marking articles as relevant. If you mark the article as relevant in the map it will get highlighted – the same goes for highlighting in the dataset. So it’s easier to find it.
User Profile
Now you can edit your information in the user profile.
Projects
We saved the best for last! The newest version of the RW has project space implemented. “Projects” is a collaborative space for your organization, where you can work together.
Keep an eye out for the next product announcement, the Researcher Workspace 1.0, which is coming out very very soon!
The European Commission judges identified Iris.ai as having the potential to support European scientific research and AI development against a growing competitive landscape coming from the US and China
London – 11 January 2023– Iris.ai, provider of a world-leading and award-winning AI engine for scientific text understanding, has been selected by the European Innovation Council to receive €2.4 million of funding in grants and up to €12M in equity investments from the 2022 EIC Fund. Iris.ai was one of just 15% of startups chosen with a female CEO and amongst only 78 successful companies from over 1,000 qualified candidates in a highly competitive selection process.
The European Innovation Council (EIC) is a flagship European program to identify, develop, and scale up breakthrough technologies. In its third year as a full-fledged EU fund, the EIC has a budget of €470 million to fund the most promising technologies across Europe and support European growth.
Iris.ai solves a major problem for researchers – that of the overwhelming volume of research being published. Finding relevant research is like finding a needle in a haystack, and as a result, researchers in both academia and industry are missing relevant published papers that could advance their knowledge or are simply wasting time reading irrelevant research. Iris.ai cuts the time required to carry out scientific research by using AI language models to categorize, navigate, summarize and systematize data from academic papers, patents and all other technical or research documentations. It’s already being used by hundreds of universities and companies – including Fortune 500 steelmaker ArcelorMittal – to eliminate the weeks and months spent manually wading through patents and research.
Iris.ai’ technology has the potential to enable unprecedented breakthroughs in European research through interdisciplinary scientific discovery against growing competition from the US and China. The EIC Jury responsible for reviewing Iris.ai’s application commented: “A strong AI/ML European player is of utmost importance for the EU, given the evident progress in US and Chinese-driven AI developments and the potential biases that could therefore be implemented in the algorithms.”
Iris.ai will use the investment for its core mission: making scientific research more actionable. As a fully horizontal platform, this funding will assist the company in accelerating interdisciplinary research across every field, mapping out the vast amount of research available and helping to broaden researchers’ understanding beyond their specific domains.
The EIC Jury concluded: “We believe that Iris.ai has huge potential, with the investment enabling them to benefit from ‘the horizontal’ competitive advantages of the product and expand into other market segments in the mid and longer term.”
Anita Schjøll Abildgaard, CEO and Co-founder of Iris.ai, commented, “When Iris.ai was founded in 2015, few people had heard of language models. Since then, the AI ecosystem has grown exponentially, and the concept of language models is common knowledge.
“However, the current generation of large language models – including ChatGPT – don’t work for science today. They hallucinate, generate mistruths, and misunderstand scientific text due to a lack of domain-specific knowledge. What we’re doing differently is we are working on factuality validation, and injection of externally validated knowledge – creating a trustworthy system that can be relied upon for analyzing scientific research. Our research efforts are considerable, compared to other startups in our field.
“Together with my two brilliant co-founders Victor Botev and Jacobo Elosua, we are delighted to be selected for the European Innovation Council Accelerator funding. It will allow us to ramp up the development of our technology and achieve our goal of building a complete AI researcher – AI tools and applications which allow humans to make sense of the totality of the world’s scientific knowledge.”
The internet has revolutionized our lives, giving us instant access to vast amounts of information. The exponential increase of knowledge together with technology improvements contributed to the creation of the Open Science movement. Artificial Intelligence has a chance to help researchers with processing the abundance of knowledge available today.
Open Science movement
Open Science consists of scholarly communications (Open Access, OA), research data (Open Data), and the protocols and other software tools that gather and process the data (Open Source). It aims to make all scientific products generated by research activities (e.g. scientific literature, research data, software, experiments) available for everyone as soon as possible. In recent years, it has become accepted that open access is a desirable and viable publication model for papers. Open Access benefits researchers, institutions, nations and society as a whole. Today, Open Access journals not only rank among the world’s largest in terms of numbers of articles published, but also among the most influential in terms of citations received. But the more the Open Science movement grows with increasingly large amounts of data available, the more complicated it becomes to utilize it effectively. In this article we will focus on scholarly articles – Open Access.
Artificial Intelligence to the rescue
Digitization of knowledge is the future of science. Artificial Intelligence has great potential and opportunities for making this knowledge overflow more manageable. More and more researchers are looking for solutions to find the right articles fast and analyze them. Searching for hours or days for the right documents, analyzing them, extracting the data needed… it all can be overwhelming and incredibly time consuming. While humans can get tired from reading large amounts of information, the machine can faster, better and without losing focus.
AI offers the promise of finding and transforming the data into valuable insights. The modern technologies have the potential to make high quality materials readily discoverable even when they would be challenging to find through the traditional distribution channels.
More and more tools are being developed to help researchers discover previously hidden information.
Our Approach
At Iris.ai, we are a big advocate for making all the world’s knowledge available to everyone. We believe that there’s a great deal of undiscovered and untapped information hidden in these articles. We believe that if one human could sit down, read all of the scientific knowledge available and draw conclusions, we would solve a lot of problems right then and there. Maybe a biologist discovered a plant with certain characteristics that could be used as an alternative to plastic or help in alternative medicines? Our goal is to help researchers make sense of existing knowledge fast and effectively.
Iris.ai Researcher Workspace is an AI-based platform for document processing.
It provides smart search and a wide range of smart filters, reading list analysis, auto-generated summaries, autonomous extraction, and systematizing of data. It’s an invaluable companion to researchers, with the ability to do inference and find new solutions from a vast body of scientific knowledge. Iris.ai allows humans to focus on value creation by saving 75% of a researcher’s time, doing specialized, interdisciplinary field analysis to an above human level of accuracy.
The Researcher Workspace has integrated multiple databases:
>100 million open access papers via core.ac.uk,
Open Access and Paywalled papers from PubMed,
USPTO patents via PatentsView
CORD-19, the Allen Institute hosted COVID-19 dataset
The researchers can search through these databases with their own problem statement and the machine will find the most relevant articles.
Unfortunately there are still many paywalled databases which we cannot integrate in our tools, but now researchers can upload their own dataset either from an external folder or from their reference manager.
Amongst our clients we work with a Californian research group, funded by the Canadian government, who is searching broadly for new drug candidates from western, Chinese and herbal medicine for a specific medical situation or we work with a start-up extracting data from documents to find candidates for creating sustainably-sourced biomaterials, made from locally-abundant natural ingredients.
The Open Science movement together with emerging new technologies to manage knowledge are the future to scientific development and new discoveries.
Key Takeaways
👉 Open Science consists of Open Access, Open Data and Open Source
👉 Artificial Intelligence has great potential and opportunities for making this knowledge overflow more manageable.
👉 Iris.ai tools help researchers save time on finding the right articles, analyze the list, summarize the findings and extract the data.
Want to know more about the Researcher Workspace? Get in touch!
We are super happy to announce our latest product launch: The Researcher Workspace 0.5! Besides the usual bug fixes, it includes some exciting new features:
👉 Adding content via Collection input
👉 Push Notification System (PNS)
👉 Bookmarking documents
👉 Document sorting by context filter relevance
👉 Removing documents from datasets
We hope you enjoy the new updates! ✌️
Do you want to try them out? Request a demo! (Currently only available for businesses)
Now you can upload your document collection to the Researcher Workspace. Whether you collected your documents in a spreadsheet with links (.csv), your reference manager (.bib or .xml).
Push Notification System (PNS)
When you are running the tools some things take time – and we understand that! That’s why another new feature is notifications! If you’re generating a map, creating a new summary, running the analyze tool or extracting data, these processes will be running in the background. So you can carry on with the work and we’ll notify you when it’s done.
Bookmarking documents
Let’s say you created multiple datasets and there are some articles that you want to get deeper into, read through it but you don’t have time now or you read them and you always want to have them at hand. Now you can bookmark them! Bookmarked articles have their own separate place which you can access through the menu bar on the left hand side.
Document sorting by context filter relevance
In the newest version of the Researcher Workspace you can sort the articles from your dataset not only by the published date, but also by title or by relevance of your applied context filters. The latter feature expands on the very cool context filters that are accessible in the Filter tool. Once you have created and applied a context filter, you receive information about each paper in the dataset and their corresponding relevance compared to it. The higher the percentage, the more similar the papers will be to the context filter. Now, you can use this information to a higher extent by sorting the article list to bring out the articles with highest relevance to you. You can use a few at the same time by marking which one is a priority. For example, if you have two context filters applied and you have selected sorting by both context filters, the system needs to know which one is a priority in front of the other. It is most prominent when papers have results higher on one of the filters and lower on the other.
Removing documents from datasets
Sometimes the filters are just not enough and you want to have more options in editing your perfect reading list. That’s why we have added the “Delete” option to the documents so if the article you’re looking for is just not it, you can simply delete it from the list with one click.
Pharmacovigilance is the science and activities relating to the detection, assessment, understanding and prevention of adverse effects with pharmaceutical products. This surveillance starts once the drug is approved for market use, after successful clinical trial completion. Though PV processes can vary across organizations, there are certain requirements that are universal, such as signal detection, risk management, and adverse event reporting. Some of these processes often take up a significant amount of time and resources and are repetitive and manual.
A Deloitte survey shows that more and more pharma companies are investing in automation solutions for their pharmacovigilance processes. According to a study, 90% of companies’ goal is to slash case processing costs by automating and implementing AI in the process.
How is AI applied in PV?
Literature screening is one of the most time consuming parts of the PV process. It involves identifying articles that contain information about serious and non-serious adverse events associated with a particular medicinal product. However, doing this manually is extremely time consuming and labor intensive. Pharmaceutical companies struggle with retrieving and analyzing scientific and medical literature. One challenge they face is developing a good search strategy. Another is understanding and extracting information from the unstructured content found within the literature. Finally, they must convert the content into adverse event data so that it can be loaded into a safety database. Automating these processes with AI – which is only now becoming possible, thanks to improvements in NLP technology – can save them time and money, reduce the workload, increase efficiency and cut out human error. For example, AI can automatically identify the most relevant adverse events and extract and clean that data.
Iris.ai Approach
With our Iris.ai tools you can automatically find the documents you need from any written text on the topic, such as research papers, patents, case reports, clinical trial reports, etc and extract all relevant information. Freeing up this time will enable PV professionals to spend time on more valuable-added tasks.
Problem
Our clients used to spend a lot of time on systematic review tasks. Diligence, attention to detail, and rigor is key, yet even at full focus human accuracy in these tasks is measured to 86%. Spending 80 hours narrowing down a reading list of 500 documents to the 40 key relevant ones, extracting all the relevant clinical data, writing a summary and a report – and repeating the full exercise on a yearly basis – is a time drain. That time is essential in preventing any more severe outcomes from adverse events.
Solution
With the Iris.ai Workspace, our clients can analyze the document set for entities (disease/drug/device), clinical data points, and contexts of interest and rapidly filter the sets down to the most relevant documents. Clinical data extraction is fully automated, populating orderly excel sheets with data from documents in seconds. Summaries for kicking off the meta-analysis are written by the machine. The review can be automated at regular intervals, and a compliance report generated.
Benefits
The Iris.ai tools can reduce the time for a systematic review for compliance or pharmacovigilance by up to 75%. The accuracy is on par with human performance (>85%). Contractors can take on more assignments at higher margins, while pharma companies can reduce the time they spend on these tasks.
Benefits of PV automation
As our customers have confirmed – in pharmacovigilance, partial automation can eliminate human error, reduce costs and time spent on paperwork. Automating pharmacovigilance helps pharmaceutical companies comply with drug safety regulations, raise reporting volumes and speed up the process.
Key Takeaways
In recent years, there has been a lot of discussion about how to effectively use AI for pharmacovigilance. It can be used for literature screening which includes identifying abstracts and articles that contain information about adverse effects and extracting them. AI can save them time and money, reduce the workload and increase efficiency.
We get loads of questions from our customers who are very curious about our new tool – the Researcher Workspace. That’s why we’ve collected some of these questions in an old-fashioned frequently asked questions.
If you have any other questions or would like a demo, we’re only one click away.
How is the Researcher Workspace different from other tools available online?
The Researcher Workspace is an AI-based research platform. What differentiates it from other tools available online is the all-in-one approach. The researcher or a science enthusiast can start an interdisciplinary search, find broad lists of articles, filter the most relevant ones, summarize their findings and lastly extract the data points – all of that using one platform! Moreover, the platform is based on artificial intelligence models, which means that the machine really understands what you are looking for and it is able to find scientific articles that other keyword-based search engines are not.
How should I get started with the tools?
Using completely new tools might be an overwhelming experience and a challenging process – we understand that! Therefore, when your business acquires the license your researchers will receive several training sessions, help materials with all steps and FAQs as well as video tutorials. For the individual users – don’t think we have forgotten about you, stay tuned for the official release in the fall.
After adding the content, the Researcher Workspace tools are interchangeable – you can combine them in any way you want. For example, you can start with an open repository dataset and then Explore the documents within by using your problem statement or link to relevant articles or you can upload your own dataset and use the smart filters to narrow down the list. Every research process is different and you can use the tools however it makes the most sense to you.
What types of content do the tools work with?
The tools work with documents of scientific language. This can stretch to trade magazines, popular science, regulatory documentation and internal memos, but the further away from ‘pure’ science we get, the less accurate the tools will get.
The tools work with almost any type of machine readable content:
PDF
As a starting point, patents and research papers in PDF formats are the primary source.
Any PDF document including scientific text and tables of data, so long as the PDF is encoded in any of the (too many) main approaches used.
Word processing: .txt, word, powerpoint etc
We can support a range of other formats, so long as they contain obtainable scientific text and/or data.
Coming soon! Collections: BibTex, RIS and CSV ++
How does the machine know which points to extract?
One of the configuration parameters for the tool is what we call the output data layout. As a client, you specify the data you’d like to extract and together with our team we will create this layout, which then gets implemented and ready to use within the Extract tool.
How do I train the machine in my research domain?
It starts with the client providing some research articles or patents describing his field or specific problem. Then we are using our engine to find thousands of related papers based on each article provided. Then we “inject” found articles into our general model to enhance the relationship of the words that are directly associated to the domain knowledge requested by the client. As a result, our engine becomes more sensitive towards domain-specific terms and their relationship to other words that are already in the model. Feedback loop with you, the domain expert, for validation of initial results.
What is the difference between an abstract and your summary?
There are broadly two approaches to automatic text summarization: extractive and abstractive. Extractive approaches select passages from the source text, then arrange them to form a summary. You might think of these approaches as a highlighter. The summarization that Iris.ai uses is abstractive summarization. This means the tool captures the context of the text and in essence, writes its own summary – it does not just copy-paste parts of the texts together, which is called extractive summaries. The machine-generated summary of Iris.ai thus contains new phrases and sentences that may not appear in the source text.
The Iris.ai’s generated summary takes out the most important parts of the article and unlike the article’s abstract, it is not biased by the authors.
What does it mean that the tool is AI-based?
Artificial Intelligence is a crucial and helpful part of the Researcher Workspace. For example, in the Explore tool, the machine extracts the most meaning-bearing words from your article or problem statement and then enriches them with contextual synonyms and topic words to build a contextual “fingerprint” which is matched with the content collection or database of scientific text the tool is connected to. Using the same “fingerprint” approach in the Extract tool, the machine identifies various values and entities across text, tables, and graphs (!) in the documents, which it extracts neatly. Moreover, the Extract tool has no problems with handling abbreviations, different measuring units, and converting them. The AI is used as well in the Analyze tool to quickly scan the dataset and provide topic and concept lists to filter the articles.
Does the machine also understand other languages?
Currently the machine works only in English, but in specific cases we can implement a translation to other languages. It is worth mentioning that most non-English scientific articles include an English version of the abstracts as well, so our machine successfully works with them too.
Watch our quick 1 minute video to get a quick overview of the Researcher Workspace.
In the last blog post we introduced our new shiny tool – the Researcher Workspace. A place where you decide how to do your research process and apply the AI-based tools you need. In this post we want to introduce you to some concrete client use cases and explain how we set up their workspace for their unique process to address their needs.
Food safety reviews
We are working with a niche review team that needed to perform a wide range of food safety related searches to keep the population safe. The team needed a better way to do full interdisciplinary literature searches on a broad variety of topics, expanding beyond their individual areas of expertise. They have a simple Workspace with an Explore tool, where the reviewer can input a human language description of the research, without the need for comprehensive vocabulary understanding. Iris.ai uses contextual key terms, synonyms and hypernyms to build a ‘fingerprint’ it can match to other papers and presents a visual map with overview over topics and relevant research papers.
Exploratory R&D for a biotech Company
This collaboration is with a biotech company that has extensive R&D efforts in mapping out a knowledge graph of real world data (e.g any microorganism they locate in their field or lab work) that will lead to new insights for their product development. Their starting point is an entity (e.g bacteria, parasite) in a given context – and then search for all relevant knowledge, analyze all articles found, extract the key data points and summarize the findings, with the goal of connecting information in a database.
Pharmacovigilance for a Contract Research Organization
This CRO performs regular contract work where they do limited, specific and rapid literature searches for their clients across a variety of medical devices and drugs.
They are regularly undertaking post market surveillance and pharmacovigilance studies. Usually each project is about 80 hours. In order to reduce the time the team spends on the manual work, the CRO has an Iris.ai Researcher Workspace setup where they upload their search results (≈500 articles), apply a range of smart filters (entities, data points and context descriptions) to instantly filter down the list to the ≈30 articles of interest. Then the tool automatically extracts all clinical data points of relevance into a database – both to a short collection of summarizing data points and to a comprehensive 500-point collection. All actions are recorded and a final report draft is automatically generated.
Drug repurposing for an interdisciplinary research group
A Californian research group, funded by the Canadian government, is searching broadly for new drug candidates from western, Chinese and herbal medicine for a specific medical situation. Each researcher starts from an idea, doing a PICO style natural language text search and filters, across a variety of western and alternative research articles. Once an interesting approach is found, the articles are collected and all relevant clinical data is extracted and used to populate a database, made openly accessible to researchers across the world.
IP analysis for a steel conglomerate
This world-leading steel manufacturer strategically monitors filed IP to spot new market opportunities, but extracting detailed experiment data from patents is incredibly time consuming. Their Workspace allows researchers to provide the system with one or several patents, using that input to identify a range of other patents based on the content. The identified patents can then with one click be sent for extraction, where thousands of data points in text and tables are extracted, linked and mapped to their desired format.
Academic reviews
Librarians are always searching for better ways to help their students and researchers find the right literature. An academic-oriented Researcher Workspace is connected to all relevant literature sources and allows exploratory searches based on problem descriptions. The tool, with its visual search interface, is especially loved by Master and PhD candidates early in their research careers, with still-unknown topics to explore broadly.
Key takeaways
Each presented use case is distinct, but in every instance we found a suitable process to help companies, research institutes and universities with their scientific knowledge processing needs. Our tools are customized and trained on each client’s domain to optimize results. If you think that we can help you out, contact us!