Training resources

If you prefer videos, we suggest beginning with one of these intro videos, which walks you through some overall scenarios:

Literature reviews with the Researcher Workspace

Read Transcript

Getting started with a new research project can be daunting. Theses, dissertations, papers, research grants - all require a lot of time spent reviewing scientific literature and discovering new knowledge to shape your ideas. Yes, it's a long way to write the first lines, especially as we're drowning in an ever-increasing number of published research. It all seems to be never ending.

A comprehensive literature review is THE FOUNDATION of your research project. Conducting a literature review is about making connections with the related works and with your own research. It is simply impossible to search and read "everything" that has been written and published on a topic. It can be an overwhelming task. But it doesn't have to be that way. Iris.ai’s research assistant is here to help you speed up the process and manage it successfully!

Iris.ai Researcher Workspace is an AI based platform with tools that accelerate your literature discovery process. It includes five modules - Explore, Analyze, Filter, Summarize and Extract. It is a new toolbox for students and researchers who want to process scientific knowledge more effectively.

Let’s take a look at how it works!

This is Alice, who is passionate about growing chili peppers. What’s more, today she’s doing a study about capsaicin, a key chemical compound that triggers that sensation of extreme heat. Yes, we all know that feeling. However, Alice isn’t interested in culinary ingredients, but in the clinical applicability of capsaicin in treating pain. The Researcher Workspace allows her to explore her topic broadly, discover new intersections, filter her reading lists and create auto-generated summaries of her findings. There are many ways on how she can go about her topic.

First, she wants to make new discoveries. That’s why, Alice does not want to be limited by keywords, which are great if you know exactly what you’re looking for, but rather to frame her ideas through context and interdisciplinary research. Moreover, none of us is an expert in all scientific fields and terminology. Not all of us speak “library”. So, Alice has created an Open Access dataset that contains millions of scientific papers and then generated an explorative map based on the article she’s interested in. The Explore tool reads the paper title and abstract and finds related articles based on the context.

Content visualization helps her uncover otherwise hidden connections and identify spot on articles! Look at this one. The machine is smart enough to indicate the relevancy score to her initial input text! To get a good overview of the topic and see it from multiple perspectives, she will easily create more maps based on other interesting articles. And of course, she can also generate a map and reading list based on her own description of her research topic, hypotheses, etc.

Alright, Alice now wants the machine to read the reading lists she has created on her behalf. What do these articles have in common, what are their key concepts? Let the Analyze tool do the job and tell us about it!

And here we go! Alice can choose to include relevant Topics and concentrate on reading these articles first.

OR she can filter the list with word analysis and choose to include the key concepts.

Lastly, she can summarize the articles that interest her to get a solid overview of the topic and to kickstart her own writing. She chooses these two articles related to anti-cancer therapies and pain management. She generates machine-written summaries from multiple abstracts, full-texts or full-text itself.

Another way to go about this is that once Alice has gotten a broad overview, she wants to systematically track all articles contextually related to capsaicin and TRPV1, published in the last 7 years in PubMed. In this case, Alice wants to capture all relevant articles and fine-tune her reading lists by using a context filter. A brief description of the additional context she would like to include in the result list. E.g. a topic related to “Anticancer drugs”. Now, she can narrow down her initially huge reading list to only the most relevant and recent articles, while the machine indicates her similarity matches directly in the reading list. Something hard to imagine in traditional keyword engines. She can export this dataset to run even more granular analysis of key topics and further refine her final findings.

But what if Alice also has her own reading lists created from other resources or subscribed databases? There are not millions, but some hundreds or thousands of articles that she was unable to further distill using keyword queries. Or what if she has even her own collection of PDF articles stored on Google Drive that she needs to analyze to see what to read?

The Researcher Workspace is flexible, so Alice can easily upload her document collections both in citation formats like EndNote, BibText or CSV, as well as full-text documents.

Then she can run the Analyze tool to further define the reading list, apply metadata or context filters, and finally summarize interesting looking articles.

This literature review was no pain for Alice. Even though her topic was really HOT and PAINFUL. Alice saved a tremendous amount of time finding and summarizing new spot-on articles. Now she can use this saved time on writing conclusions she is confident about. And of course, also on growing chili peppers.

Automate your research with Iris.ai Researcher Workspace.

Medical writing with the Researcher Workspace

Read Transcript

Are you a medical writer? Do you want to speed up your literature review process and get your writing off the ground?

The demand for medical writing is growing steadily in the pharmaceutical and healthcare market. Medical writers are highly skilled professionals who can spend up to 2 months doing scientific literature review before they summarize their conclusions. It's mentally demanding job that requires lots of concentration, expertise, energy, and unfortunately mental stress when it comes to deadlines.

Iris.ai Researcher Workspace can ease this pain and help medical writers speed up their literature review process. The tool includes five modules - Explore, Analyze, Filter, Summarize and Extract - that streamline the entire medical writing workflow

This is Ben and today his task is to find relevant clinical articles on female reproductive senescence. Searching in PubMed presents him with almost 20 000 documents and even after filtering it, it leaves him with thousands - or in better cases, hundreds of articles -that would take a long time to go through. Skimming through endless reading lists, and finally extracting relevant data is the hardest part of the process.

The Researcher Workspace can be adapted to a variety of different use cases, allowing Ben to combine modules to best suit his process.

For example, Ben can start his process by creating a dataset from live databases, such as PubMed. Then explore his topic broadly and from different angles, based on his problem statement or relevant articles. Yes, the Researcher Workspace understands your problem contextually and goes beyond traditional keyword searching! The best way is to create a number of different maps to get a good grasp of the topic. After that, he can refine the findings with the context filter. The usefulness of the context filter comes when you need more than one word to describe your desired outcome. And finally, when Ben is presented with his distilled reading list, he’s two clicks away from summarizing a number of abstracts or full-texts to deepen his understanding of the topic and get his writing off the ground!

Another way he can go about it is to again create a PubMed dataset, apply the Published date filter, and use the analyze tool to choose concepts for inclusion and exclusion. Then save the filtered dataset as a new one and run the Analyze tool again to get more and more specific concepts and topics to fine-tune his reading lists. Repeat this process as many times as needed until he ends up with a short, but very granular reading list. He can simply export the list into his preferred format (such as EndNote/BibTex/CSV) or again, he can summarize interesting looking articles.

The Researcher Workspace also comes with the Extract tool that allows you to extract data from tables or text from the articles into your predefined spreadsheet. For example on baseline characteristics or adverse events.

Instead of spending weeks on literature searching, sifting through endless reading lists, and stressing over approaching deadlines, Ben can now use the Researcher Workspace to complete this work in a matter of hours and use the time saved to write instead.

Enjoy your working days. Automate your research with Iris.ai Researcher Workspace.

Transforming Engineering Research with RSpace™: A Case Study in Advanced Materials

Read Transcript

Alice, a dedicated researcher in anti-corrosive technologies for metal coatings, faces a monumental challenge in her work. Her task involves extensive prior art searches across numerous papers and patents, a process that is both time-consuming and demanding. Navigating the vast sea of data, with its complex technical jargon and hidden insights, hampers innovation and scalability, making manual extraction of detailed material properties laborious and error-prone.

The breakthrough comes with RSpace, a powerful AI-powered research assistant that transforms Alice's workflow. RSpace securely connects her documents with external research databases, enabling efficient exploration and extraction of actionable insights from complex data. This tool significantly enhances Alice's ability to map out competitive information and discover novel insights, driving her company's success in advancing anti-corrosive technologies.

When Alice discovers the RSpace, it transforms her approach towards research.. With the Analyze tool, she can quickly filter through vast amounts of research in real-time using specific concepts, and related topics. This tool helps her uncover hidden connections and novel insights that traditional searches might miss.

By first filtering with Analyze concepts, Alice can save her results as a new, narrowly focused dataset consisting of articles that delve deeply into her specific topic. This ensures that the dataset contains in-depth discussions rather than mere keyword mentions. Leveraging the power of the Smart Language Model trained in materials science, Alice can then perform further analysis to uncover additional niche topics and hidden connections within this refined dataset.

For instance, she can effortlessly find detailed information about the corrosion resistance of waterborne epoxy zinc coatings or insights on modified graphene oxide used in anti-corrosion coatings.

The Chat feature within RSpace becomes an invaluable tool for Alice. By asking a few specific questions, she can quickly obtain concise summaries and actionable insights, and also references from her dataset. This gives Alice accurate information and better control over her research, significantly saving time and effort.

Alice also leverages the Explore tool to build visual maps of her research topics, using the provided references to further explore and expand on the topics. These maps allow her to contextually dive deeper into the subject matter, visualizing connections between related topics and documents. This bird's eye view helps her validate findings and broaden her horizons across disciplines she might not be an expert in. Because RSpace is connection agnostic, Alice can easily and securely search external sources, including patents and articles, as well as her company's institutional repository, tapping into the treasure trove of accumulated knowledge that is otherwise of little use for research on their cloud drive.

Systematically extracting and linking granular information on material properties and coating characteristics from PDFs and patents into a spreadsheet becomes a seamless process with RSpace. This enables Alice to synthesize data effectively, identifying the right articles and patents without manually reviewing each document. As a result, she can focus on synthesizing knowledge rather than gathering data, staying on top of research while reducing research time and costs.

Ultimately, RSpace empowers Alice to save substantial time and effort, leading to faster innovation and more efficient patent filing. By harnessing the power of this AI-powered research assistant, Alice keeps her company ahead of the competition, driving advancements in anti-corrosive technology research with unparalleled efficiency. The AI-driven insights and streamlined workflow provided by RSpace enable her to navigate the complexities of her field, turning challenges into opportunities for groundbreaking discoveries.

Food Science Research with Iris.ai Researcher Workspace

Read Transcript

Using technology and scientific principles from multiple fields such as biology and chemistry, food scientists can make food easier to grow, safer, and healthier. An in-depth literature review is crucial in this quest for excellence. Staying on top of research provides a significant advantage over competitors. Where traditional keyword search fails, contextual understanding with an AI-powered research assistant can revolutionize workflows. Sarah here is searching for novel ways to create and maintain crisp texture in cut potato products, such as chips! But in fact, some patents hide chips behind expressions like "a high amplitude corrugated food product". However, Sarah finds herself overwhelmed by the vast sea of research data and patent documents, struggling to discern valuable insights amidst the number of documents. There certainly isn't a lack of articles, but only a handful can be truly significant. How can she see what is really in there? What’s the best knowledge and how to distill it without spending hours figuring it out? Her ambitious goal is to conduct an extensive prior art search on existing knowledge, decipher deliberately hard-to-read patents, and collect relevant data for their lab. With the Researcher Workspace, Sarah effortlessly sifts through extensive datasets, thanks to its advanced concept filtering capabilities. She begins by narrowing down her search from a staggering 145 million articles to a focused dataset centered around crispy potatoes. Let's unravel what further niche topics and hidden connections are inherent in this dataset through in-depth analysis. Sarah quickly navigates to the Topics section that now shows the topics focused specifically on potatoes, crisp and snacks.

In the concepts section, the machine comprehends specific evaluation metrics unique to this field, such as crispness score, retention, hardness, and perception. Sarah utilizes the context filter to specify her focus on "crispness retention." She provides additional context by writing a brief explanation of what she aims to achieve and defines variables and factors influencing this such as moisture content, oil quality, effective packaging to prevent air and moisture exposure, and optimal storage conditions that are connected to crispness retention, enhancing the precision of her search. With a manageable dataset at her fingertips, Sarah engages now with the Chat tool. Armed with specific questions, she accesses concise summaries and actionable insights, cutting through the complexity of hard-to-read patents and dense research papers. With provided references Sarah creates a map with relevant articles to broaden her understanding and uncover new connections within her research domain.

As Sarah delves deeper into her research, she utilizes the Extract tool to extract pertinent data from selected documents effortlessly. This feature streamlines her prior art search, enabling her to decipher and collect relevant data on crispness properties for her lab experiments.

Moreover, the Researcher Workspace offers custom project-based extractions tailored to Sarah's specific research needs. These tailored layouts include only the data points relevant to her study, such as texture-enhancing ingredients, processing techniques, and sensory analysis results. With the click of a button, Sarah can extract and link together all the critical information scattered across documents, empowering her with comprehensive insights for product development.

Excited about her newfound discovery, Sarah eagerly shares the Researcher Workspace with her colleague, eager to demonstrate its capabilities and streamline their research efforts together. By staying ahead of the curve and swiftly acquiring crucial data, she gains a significant advantage over the competition, propelling their team towards success.

Advancing Biotech Research with Iris.ai Researcher Workspace

Read Transcript

Meet Clare, a dedicated researcher in the quest for novel, plant-derived therapeutic drugs to effectively combat cancer. Her ambitious goal is to develop clinically useful small-molecule drugs with a foundation rooted in the rich source of polyphenols found in plants.

Clare's struggle lies in drowning in data collection rather than synthesizing knowledge for novel therapeutic drug research. Traditional keyword searches fail to provide the depth she needs in her niche area, leaving her sifting through an expanding corpus of research. Despite her efforts, Clare grapples with the intricate language and complex concepts that obscure critical information.

But Clare's journey takes a transformative turn when she discovers the AI-powered Researcher Workspace.

The Analyze tool allows her to quickly filter through vast amounts of research in real time using specific concepts also covering synonyms, and topics related to plant-derived therapeutic drugs and polyphenols. It helps her uncover hidden connections and novel insights that traditional searches might miss.

The Chat tool within the Researcher Workspace becomes Clare's trusted companion. With just a few specific questions, she can access concise summaries and actionable insights, along with real references from her dataset. This empowers Clare with factually correct information and control over her research inputs and outputs, saving her valuable time and effort.

She can leverage provided references to build Explore maps to contextually dive deeper into the topic. The map visualizes connections between related topics and documents. This bird's eye view helps her validate findings and broaden her horizons across disciplines in she might not be an expert. Clare strengthens her research and uncovers new potential avenues for innovation in the development of plant-based cancer therapeutics.

Once Clare has narrowed down her research, she utilizes the Extract tool to extract data from selected documents with just a few clicks. This feature saves her an enormous amount of time and allows her to delve deeper into the data to extract valuable insights for her research.

Moreover, the Researcher Workspace offers custom project-based extractions tailored to Clare's specific research needs. These layouts include only the data points relevant to her studies, such as types of polyphenols, proteins, plant-derived molecules and their potential therapeutic effects. The machine will extract and link together all the properties scattered in text, tables, or chemical formulas in the friction of time. This customization enables Clare to focus precisely on what matters most for her research goals.

Clare develops greater confidence in her research methods as she explores the Researcher Workspace. She uncovers new possibilities in cancer therapeutic drug research with every query and click while being automatically alerted to the latest research. . This brings her closer to her team’s goal of developing effective and innovative therapeutics. Her boss realizes that staying on top of research gives them a huge leg up on the competition.

Advancing agrochemistry research with Iris.ai Researcher Workspace

Read Transcript

Meet Grace, an agriculture researcher striving to find innovative solutions for efficient nitrogen fertilizer management. As she delves deeper into her field, she encounters a myriad of challenges, from the overwhelming volume of research documents to the complexity of optimizing nitrogen usage.

Amidst the sea of information, Grace finds herself lost, struggling to navigate through the maze of data and studies. It's like searching for a needle in a haystack, with each document offering a piece of the puzzle but lacking cohesion. How can she stay on top of her research?

Grace's journey takes a turn when she discovers the Researcher Workspace, a secure, fact-based AI research assistant.

Within the Researcher Workspace is the Explore tool, a content based search engine that empowers Grace to uncover hidden insights and novel connections both within her research domain and cross-disciplinarily. With each click, she embarks on a journey of exploration, navigating through contextual topics like nitrogen management, organic fertilizers, and formulation techniques.

She utilizes the Analyze tool with Concept filters, which prove invaluable in quickly filtering the list of results with not only specific keywords but also related synonyms and topics, otherwise missed by traditional search engines.

But what truly excites Grace is the Chat tool, her trusted companion in unraveling the newest discoveries in the use of nitrogen fertilizers. With just a few specific questions, the tool delivers concise summaries and actionable insights, along with real references from her dataset. Factually correct information goes hand in hand with Grace's control over both inputs and outputs.

After exploring and filtering, Grace is left with a short list of documents. Now she can extract the table data with just a few clicks using the Extract tool. This allows her to save an enormous amount of time and delve deeper into the data to extract valuable insights for her research.

The Researcher Workspace is also able to extract data into custom Output Data Layouts. These are specially designed spreadsheets that include only the data points relevant to Grace's research, such as fertilizer properties, molecules, and emissions. This customization allows Grace to focus precisely on what matters most for her studies.

As Grace explores the capabilities of the Iris AI Researcher Workspace, she gains confidence in her research methods. With each query and click, she opens up new possibilities in agricultural research. This efficient approach saves time and paves the way for more sustainable farming practices, allowing Grace to focus on innovation rather than just tedious data collection.

Patent analysis with Iris.ai Researcher Workspace

Read Transcript

Prior art search is important if you’re thinking about competitive knowledge or securing your new invention. It is the only way to determine whether your invention has been publicly disclosed. All this information is available as state-of-the-art - everything disclosed to the public, including both patents and non patent literature.

With Iris.ai Researcher Workspace researchers can easily find related patents and articles, filter the results, export the list, summarize findings and extract key data. All of this within a few minutes instead of weeks. Iris.ai Researcher Workspace includes modules like: Explore, Analyze, Filter, Summarize and Extract.

Here’s Adam. His task is to do prior art searches on a daily basis. His task now is to identify similar content to “Multivitamin-mineral regimens for longevity and wellness” and map out the landscape of literature.

The Researcher Workspace can be connected to both internal documents (like PDF collections) or external collections (paywalled or Open Access). Iris.ai has an integrated USPTO (United States Patent and Trademark Office) database so it’s perfect for patent search.

Adam begins his search by exploring all related documents to a patent he’s interested in.

The researcher workspace creates a map where relevant articles are divided by topics. After creating about 10 of these maps, he has a good overview over all aspects of this field.

After collecting all these patents in one dataset, he can begin filtering using the machine’s analysis and context filters.

Look how quickly he ended up with a short and relevant list!

Adam can now export the results and continue working on them OR He can choose the patents and extract the data from them. Here’s how it looks.

The tool comes with a self-assessment module, which reports how confident the system is about each extracted data point, so you can easily access the results. All of this is done up to 90% accuracy - at least as good as human performance.

This extraction was still easy. Let's have a look at a more complicated one with a lot of variables to collect and link. Here is how other examples can look like.

So, Adam has managed to contextually identify the similar patents and extract all relevant data at a fraction of the time, which made his day brighter and lighter. And his company can expedite innovation based on his research outcome.

Iris.ai Researcher Workspace helps you make sense of your research. Saving time and money.

Enjoy your working days. Automate your research with Iris.ai Researcher Workspace.

Data Extraction for Pharmacovigilance & Regulatory Affairs | Iris.ai

Read Transcript

Another day at work and you spend your time analyzing an endless list of articles trying to extract all the data. I bet you’re tired!

Yes, pharmacovigilance or compliance-related research workflows are tedious. Everyone knows that. They require a lot of time and concentration, are repetitive, exhausting and thus error-prone. But they are so indispensable for securing patient safety and driving innovation at every pharma, or medtech company. Ooh, what a responsibility to do this job right!

But our technology comes to the rescue…

Imagine doing months of manual, dull tasks within minutes! You need to extract all adverse effects of sildenafil from 1 000 articles? No problem. A few clicks on a computer, make yourself some coffee and when you’re back the extraction is done. And it’s not just a dream - it’s a reality.

Michelle, right here, spent on average 10-15 hours to identify the right clinical content, narrow down the result list and extract all relevant data on adverse effects just from one document. After automating this workflow by the machine, she is now able to complete this laborious task within minutes and yes, including systematized data extractions from multiple documents!

So instead of wasting her time collecting data, she can now concentrate on analyzing the results while enjoying her coffee.

Iris.ai’s technology can save you time and money, reduce the workload, increase efficiency and cut out human error - and detecting adverse events faster and easier can save more people.

The Extract tool is able to extract, as the name suggests, all relevant data from text, tables and graphs - connecting it all together. The tool is a part of Iris.ai Researcher Workspace which includes modules like: Explore, Analyze, Filter, Summarize and Extract. If your desired data layout is accepted and integrated in a tool you can extract the paper only with a few clicks.

All you need to do is to upload your documents to the Researcher Workspace, mark which ones you want to extract, choose your ODL (output data layout) and let the extraction begin.

Enjoy your working days. Automate your research with Iris.ai’s Researcher Workspace.

We also recommend checking our playlists with:

Training resources ​

Training resources