Author: Nick Brown

Identifying Document ID & Data Source in Vyasa Layar

Identifying a document ID or data source is useful for both search and API calls. Below is a step-by-step guide for identifying both via the Layar interface.

Identifying a Document ID

1. Select a document from the Layar data catalog.

2. Click the three-dot menu on the right-hand side of the Layar document view.

3. Select “Access with Layar ID” from the drop-down menu.

4. Copy/paste Layar ID from pop-up window.

Identifying a Document Data Source

1. View document URL in the web browser.

2. Identify document data source directly after Document ID in URL.

For more information, please visit the Vyasa Developer Hub.

Table Extraction Tutorial

How to use Vyasa Layar’s automated table extraction feature.

Step 1: Click the “View Original” menu item from the Three Dot menu in the upper left hand corner of the Document view.

Step 2: Navigate to the page with the table you are interested in extracting. Click and Drag a box to fit around the table.

Step 3: Click “Extract Table”

Step 4: Click out of the PDF Preview and refresh your webpage.

Step 5: Click the “Related Documents” link beneath the title of the document.

Step 6: Use the filters on the right-hand side and select “Tables”. If you’ve extracted multiple tables, you’ll need to search through the thumbnails to identify which one you are interested in. Click the thumbnail of the table you are interested in.

Step 7: Click “View Detail” to view the extracted table in a Document View in Layar.

Step 8: Compare to the original PDF table by clicking “Compare to Original Image” on the right-hand side of the table.

Step 9: Make any touch-ups or edits necessary for the table. All edits are saved automatically.

Scaling AI in Healthcare & Life Science with Intelligent Data Fabrics – Vyasa at SC22 with Oracle, NVIDIA

According to IDC’s AI InfrastructureView Survey, only one-third of companies surveyed claim to have reached mature AI adoption. The survey notes that infrastructure remains the most consequential, but least mature decision when deploying AI. 

In most cases, companies build custom environments for deploying AI, all of which require massive amounts of data for their workflows to operate. Unfortunately, this scenario poses challenges by deepening data silos and creating redundant or obsolete data as AI deployments become mature. Over time, this negatively impacts the quality of data teams have access to as well as the productivity and enhanced intelligence companies are looking to achieve through these projects.

A Paradigm Shift in Data Analytics

Innovations in data management and deep learning are addressing this issue and presenting a paradigm shift in how companies access and apply analytics to their data. Known as a data fabric, companies can now overcome silos by making data accessible and searchable in a single platform. 

At Vyasa, we’ve combined this approach with advanced deep learning to create our Layar intelligent data fabric. With Layar, companies can connect to disparate data sources regardless of storage location. Layar then applies transformer-based deep learning to the connected data sources, making the content easily searchable and accessible all in natural language. This enables Layar to expand its understanding of both structured and unstructured data, allowing companies to gain new insight and value across their data landscape. 

As a result, companies can scale AI deployments by applying deep learning models across silos via the Layar data fabric.

Accelerated Analytics with Vyasa, Oracle and NVIDIA

Innovative cloud infrastructure from Oracle leveraging NVIDIA accelerated computing provides a high-performance environment for deploying a Layar data fabric.

As an Independent Software Vendor partner in the NVIDIA Partner Network (NPN)  ISV Partner, Vyasa accelerates these capabilities with the NVIDIA AI Enterprise software suite, which includes essential AI workflow processing engines from NVIDIA, including transformer-based deep learning models like NeMo Megatron, NVIDIA Triton Inference Server, NVIDIA RAPIDS data science libraries and more.

Deep Learning Hits its Stride in Healthcare 

Healthcare technology leaders are prioritizing data integration, NLP and data annotation in their AI investments, according to recent reports from Gartner. The Layar data fabric provides a suite of low-code application interfaces that meets many of these priorities, enabling healthcare and life science teams to:

  • Accelerate early research and systematic literature review
  • Enhance clinical trial assessment and enrollment
  • Complement small compound analysis 
  • Improve target discovery
  • Better analyze genomics content
  • Streamline analyses of biomedical images

With Layar, users have experienced a 97% improvement in query accuracy for clinical trial protocol assessment and a 25% increase in productivity when compared to manual literature review.

Vyasa, Oracle and NVIDIA at SC22 

Interested in learning more about Vyasa’s deep learning analytics for healthcare & life science? Visit our demo in the Oracle & NVIDIA booth #2406 at SC22 from November 14-17. 

Access a free trial of Vyasa here



85% of clinical trials end with less than expected enrollment, while 19% are terminated or fail to meet accrual goals at all. This scenario has significant impact on the viability of trials, leading to insufficient results, loss of time and financial resources and ultimately a reduction in innovation.


One of the biggest challenges associated with clinical trial enrollment is identifying candidates due to the vast amounts of siloed and unstructured data needed. In fact, over half of eligibility criteria is unstructured, making it incredibly complex to search and access. Attempting to make sense of this content, combined with searching across data silos such as patient records and genomics data is incredibly time intensive, with key insights and candidates missed.


A leading U.S. cancer research center is turning to Vyasa to solve this problem. By establishing an intelligent data fabric with Vyasa Layar, the research center has created a unified data platform to search and identify relationships between clinical trials and its patient population to improve trial enrollment and viability. With Vyasa, the research center can:

  • Unify siloed data sources including patient records, genomics files and clinical trials into a single platform without moving or replicating files.
  • Search across multiple data sources and quickly identify insights with the power of deep learning.
  • Explore data in natural language. No coding or data science experience required.
  • Discover relationships in their data through dynamic knowledge graphs.
  • Visualize their data in smart spreadsheets and real-time dashboards.

With Vyasa, organizations empower users across skill sets to leverage deep learning and gain new value from their data. In turn, users have experienced a 25% improvement in productivity around data analysis tasks, all without compromising on accuracy.



The NIH’s Childhood Cancer Data Initiative is set to revolutionize the data landscape around pediatric cancers and treatments. While important progress is being made, hospitals and oncology departments still face an uphill battle with the resources they have available to understand patient illnesses and gain the insights they need.


The healthcare industry faces a number of data challenges, all of which are impacting pediatric oncology. These include:

  • Data silos across departments and institutions.
  • Limited access to the latest research and trials on pediatric cancer and treatments.
  • Tools for quickly accessing insights in complex data. Ex: published research, biomedical images, drug-like compounds, genomics files.


Advancements in deep learning combined with a novel data architecture developed by Vyasa are changing the way hospitals, researchers and oncologists can collect and analyze data related to pediatric cancer. Known as an intelligent data fabric, Vyasa enables organizations to unify disparate data sources regardless of cloud or on-premise storage location and without moving or replicating the data. Once connected, Vyasa applies deep learning on the data making it easily searchable and accessible. With an intelligent data fabric, users improve search accuracy by 97% while decreasing manual analysis time by 25% through the ability to:

  • Collaborate and share data across departments, and identify trends across patient populations.
  • Easily search and extract insights from genomics data, including processing VCF and BAM files.
  • Harmonize multiple data sources together, including structured and unstructured content. (ex: scientific literature & images).
  • Discover terms and insights not listed in ontologies with Vyasa’s named entity recognition (NER).

The Vyasa intelligent data fabric comes with a suite of low code interfaces that enable users to search and visualize their data in natural language, accellerating access to insights for physicians, oncologists, researchers and more.

Press Release: Vyasa Adds Real-Time Dashboards to Layar Data Fabric 

New application interface provides advanced visualization for identifying trends and anomalies across your data 

BOSTON, September 27, 2022Vyasa, an innovative provider of highly scalable deep learning A.I. analytics software for healthcare, life sciences and business applications, today introduces its latest application interface, Signal. Featuring an intuitive design, Signal enables users to monitor trends and identify anomalies in their Layar data fabric through highly-visual charts and graphs, delivered in a single dashboard.  

According to industry analysts, between 80-90% of organizational data created today is unstructured. While rich in insight, the complexity of this data makes it increasingly difficult to analyze efficiently, requiring many to rely on manual processes to find the information they need. Vyasa leverages proprietary deep learning models to accelerate this process, allowing users to search and identify insights from their data via its Layar intelligent data fabric. Signal is the latest application interface developed by Vyasa to make this data easy to analyze in low code. 

Users of Signal are able to:  

  • Visualize term usage and frequency in your data via line, bar and pie charts. 
  • Monitor terms and concepts of interest across multiple cloud and on-premise data sources without having to move or replicate files. 
  • Discover common concepts that relate to your terms of interest. 
  • Identify emerging terms and trends with named entity recognition. 
  • Receive real-time alerts to events occurring in your data. 
  • Filter charts and graphs to receive dynamic results for comparisons and reporting. 
  • Query across multiple data sources and file types. 

“Today’s organizations face significant data accessibility challenges – from content being stored across multiple silos to managing various structured and unstructured file formats to outdated processes for collecting and analyzing data that are time and resource intensive,” said Vyasa Founder & CEO, Dr. Christopher Bouton. “At Vyasa we’re developing cutting-edge deep learning technologies to address this issue and Signal is the latest example of that. Now users can quickly access the insights they need in charts and graphs which are familiar and easy to analyze without having to worry about finding the data or having the technical acumen to develop a code or query to collect the information they’re seeking.” 

Signal is being used across a variety of industries to help users to monitor trends, identify needs in the market, discover emerging technologies, streamline competitor intelligence and more. It joins Vyasa’s suite of deep learning applications powered by its Layar data fabric, including Axon dynamic knowledge graph capability, Cortex data fabric creation and management, Synapse “smart table” technology, and Retina image management and analysis. Vyasa can be deployed on-premise or in the cloud as a fully containerized application. 

Access a free trial of Vyasa’s suite of deep learning applications at  
For more information, please visit or contact [email protected]   

About Vyasa: 

Founded in 2016, Vyasa Analytics provides highly-scalable deep learning software and analytics for healthcare, life sciences and business applications. Vyasa’s technologies enable organizations to integrate and access data across disparate silos, regardless of location or structure. With Vyasa’s collection of deep learning applications, users can ask complex questions across large-scale data sets and gain critical insights to make faster, more accurate, business decisions. For more information, please visit  

Accelerating Clinical Trial Enrollment – Vyasa at HETT 2022

According to industry research, 85% of clinical trials end with less than expected enrollment, while 19% are terminated or fail to meet accrual goals at all. Considering the millions of dollars and significant time invested into these trials, the ability to identify and enroll viable candidates is key to successful trial completion and driving innovation around cutting-edge therapeutics. 

Clinical Trials’ Data Problem

One of the major challenges associated with clinical trial enrollment is the vast amounts of data required to identify relevant candidates. This data is stored across multiple silos and formats making it difficult to quickly access and analyze. In fact, over half of eligibility criteria for clinical trials is unstructured, and that’s not considering the other critical data points that help influence candidate identification such as patient records, clinician notes, genomics files, images and more. 

Traditionally, trial sponsors have relied on manual processes to identify appropriate candidates – searching through massive databases and multiple resources before gleaning out potentially relevant participants. These processes are time and labor-intensive; in many cases, only a fraction of potential leads are found. 

Tackling Your Data with Deep Learning

Deep learning is revolutionizing how the healthcare and life sciences industries can search and access data. Known as transformer-based deep learning, these advanced models use a novel approach called self-attention to train on large sets of data. Unlike traditional deep learning models that process each term separately and without context, self-attention allows transformers to build rich representations of the data to understand the relevance of the location of a term, the relation of one term to the next (even if far away from each other) and more. When trained on larger datasets, these models reach remarkable accuracy and recall for understanding the context within unstructured data like large documents of natural language text.

At Vyasa, we leverage this novel approach to deep learning as part of our Layar intelligent data fabric. By applying deep learning to disparate data sources, Layar is able to make the contents of those sources searchable and accessible in a single platform, or data fabric. We then take this a step further by providing a suite of low code application interfaces to help users simultaneously explore and gain access to key insights in their data regardless of its format or location it’s stored in. 

Successful Candidate Identification with Vyasa  

While data fabrics provide benefits across healthcare and life science, they hold significant power when applied to research tasks including clinical trial design and candidate recruitment. 

For example, healthcare organizations are turning to Vyasa to enhance their understanding of their patient population and improve clinical trial enrollment as a whole. By connecting their data silos into a single data fabric they can have all of the relevant information they need in a single platform without having to collect, move or replicate data. The addition of deep learning on top of this data enables them to uncover new insights with peak accuracy and speed. As a result, these organizations can leverage Vyasa Layar to create a unified data platform to search and identify relationships between clinical trials and their patient population to improve trial enrollment and viability. 

Working with Dell Technologies, we can provide this solution on innovative compute and storage hardware, allowing organizations to access and operate these capabilities in a familiar server environment.

Smarter Trial Analysis – HETT 2022

Vyasa will be presenting our Layar intelligent data fabric and clinical trial analysis capabilities at Dell Technologies stand #E30 from September 27-28 at HETT 2022. 

Interested in meeting? Contact us here

Access a 7-day free trial of Vyasa Layar here

Accelerating Small Compound Analysis with Vyasa and MegaMolBART

The development of a new drug takes an average of 10 years, with the discovery and preclinical stages taking three to four of those years. Considering only 1 in 5,000 drug candidates result in a commercial product, it’s astonishing to think of how much time and resources are invested by pharmaceutical companies, with relatively low success rates.

These investments are critical to the success of new therapies, from life-saving drugs to daily remedies. However, what if pharmaceutical companies had the ability to predict drug viability more accurately; starting at the molecular level?

Transformers Enter Molecule Generation 

Some of the keys to early-stage research are large-scale reviews of scientific literature and identifying relevant small molecules, or hits, that relate to a therapeutic target. Traditionally, these tasks have been conducted via manual processes, including physically reading through literature and conducting real-world experiments. 

Deep learning in the form of transformer-based models has revolutionized how researchers approach complex texts and scientific literature through advanced natural language processing. Transformers’ ability to understand the context around language structures through the use of self-attention has enabled researchers to accelerate analysis and insight discovery with peak accuracy.

While transformers have been applied against large-scale sets of unstructured text, the same approach can be applied to the structures of drug-like molecules, as represented in the “Simplified Molecular Input Line Entry System” (SMILES) format to enhance our understanding of molecules. 

MegaMolBART, a transformer model that leverages the NVIDIA BioNeMo framework, is being used to accelerate generative chemistry. Instead of the transformer training across large sets of unstructured text, it applies the same self-attention to SMILES libraries to understand and predict each individual string and its representation to a molecule. As a result, MegaMolBART can identify relationships in SMILES strings, interpolate between molecules and fuel de novo molecule generation.

Combining these two approaches means research institutions and pharmaceutical companies can revolutionize how they approach early-stage drug development, allowing for greater knowledge and more informed decisions when determining leads and trial design.

The MegaMolBART model, which is a part of the NVIDIA Clara Discovery platform, understands chemistry and can be used for a variety of cheminformatics applications in drug discovery. The embeddings from its encoder can be used as features for predictive models. Alternatively, the encoder and decoder can be used together to generate novel molecules by sampling the model’s embedding space.

A Unified Tool for Drug Discovery 

Transformer-based deep learning models have defined the art of the possible for life science and pharmaceutical companies. However, what’s often missing are the tools to bring this data together and make these models approachable.

At Vyasa, we’ve developed a novel approach for life science and pharmaceutical companies to streamline drug discovery through our Layar intelligent data fabric. By applying deep learning across an organization’s data landscape, we’re able to create an AI-powered data fabric that allows our customers to integrate and analyze their data, regardless of its storage location or file structure. These advanced deep learning models, which can be swapped in and out of the Layar data fabric given its component architecture, are used via a suite of application interfaces that accelerate insight discovery across structured, unstructured and semi-structured content — enabling users to make informed decisions for life science R&D, including the analysis of drug-like compounds. 

De Novo Discovery with Vyasa and MegaMolBART  

As an NPN ISV partner and NVIDIA Inception member, we’ve worked with NVIDIA to test and leverage advanced technologies including transformer-based deep learning models. Recently, we were given the opportunity to work with MegaMolBART to apply the model as an analytical module within our Layar data fabric. 

Our data science team focused on two use cases as part of our work with MegaMolBART — to identify similar SMILES strings and to interpolate SMILES between different molecules. As a pre-trained model, MegaMolBART could be seamlessly deployed within Vyasa, allowing the team to quickly run analytics tasks and test outcomes. 

Providing superior yields to similar models tested, we immediately identified the value of MegaMolBART. Most notably, MegaMolBART provided:

  • Significant increase in relevant molecule generation aligning with NVIDIA’s 0.989 validity score.
  • Higher frequency of unique and novel molecule development. 
  • Rapid results with GPU-enabled compute.

The implementation of MegaMolBART as an analytical module in the Layar data fabric represents tremendous benefits for organizations looking to leverage transformer-based deep learning models to fuel de novo molecule discovery within Vyasa Layar. 

The Next Age of Drug Discovery 

In an age where global health is driving demand for faster innovation around new drugs and therapies, the application of transformer-based models holds significant promise in streamlining early-stage R&D. By applying these models within intuitive, low-code applications, life science and pharmaceutical companies can improve productivity while reducing investment on low-probability programs, which leads to improved innovation and ultimately a healthier world.

Learn more about drug discovery at NVIDIA GTC, a free, global AI conference running online Sept. 19-22. 

A few can’t-miss talks include: 

1. Tuesday, Sept. 20, at 8:00 a.m. PT |  NVIDIA founder and CEO Jensen Huang’s keynote 

2. Tuesday, Sept. 20, at 11:00 a.m. PT |  NVIDIA VP of Healthcare, Kimberly Powell, special address: The Rise of AI and Digital Twins in Healthcare

3. Wednesday, Sept. 21, at 11:00 a.m. PT | NVIDIA Global Healthcare AI Startups Lead, Renee Yao, panel: Accelerate Healthcare and Life Science Innovation with Makers and Breakers

4. Wednesday, Sept. 21, at 10:00 a.m. PT | NVIDIA Clara Discovery Product Manager, Abe Stern, talk: AI-Powered Drug Discovery for Generative Chemistry and Proteins [A41196]

Boolean Operators in Layar Search

While Layar can handle keyword and boolean search, we find it helpful to provide an overview of boolean operators to enhance your Layar search experience. 

What are Boolean Operators? 

If you have ever Googled something, you have already created Boolean search strings. If you use it without realizing it, you can learn a few Boolean operators that will drastically improve your current sourcing efforts.

NOTE: Be sure to include single or double quotes to trigger the Boolean search.

Boolean Operators

  1. AND
    Narrows your search results to include only relevant results that contain your required keywords (e.g. “Alzheimer’s” AND “target” will only yield documents that have both of those keywords).
  2. OR
    Expands your search results so all results must contain at least one, if not more, of your defined keywords or phrases. OR is useful for two scenarios: (1) you need to include all synonyms for a given title, phrase or word (e.g. interleukin-1 OR IL-1 OR IL1), or (2) creating a list of all possibilities where you only need at least one of the keywords to be returned (e.g. Merck OR Pfizer OR Novartis).
  3. NOT
    Limits your search by excluding defined keywords and/or phrases from your results (e.g. drug NOT vitamin reveals all results that discuss drugs, excluding those that also mention “vitamin”).
  4. “Quotation Marks”
    Use quotes around a phrase that needs to be returned in that exact order. For example, “United States of America” will yield results with that exact term, but without quotes, each word in the phrase will be treated separately, as if you used OR between each word (United OR States OR of OR America).
  5. (Parenthesis)
    Parentheses are used to give priority to the keywords contained within over the other elements around it. As a rule of thumb, parentheses should be used around OR statements, ensuring the search engine properly resolves the OR statement before moving on to other operators. For example, (Tumor OR Carcinoma OR Cancer) AND treatment will yield results that mention treatment and any one or several of the key terms in parenthesis.
  6. Asterisks*
    * on a term matches multiple characters preceding a stem. For example, hydroxych* yields results for hydroxychloroquine, hydroxycholesterol, etc.
  7. Question Mark?
    ? on a term matches a single character preceding a stem. For example, a?duction can yield results for abduction and adduction. Another example can be anemi? to find results for anemia and anemic.

Healthcare Fireside Chats Episode 2 – Operationalizing Your Data with Deep Learning

The industry estimates that nearly 80% of organizational data is unstructured.

While rich in insight, unstructured data is incredibly challenging to search and analyze due to the complexity of content such as scientific research, scanned documents, clinician notes, pathology reports and more.

Fortunately, advancements in deep learning are creating a paradigm shift in how we can approach this data through transformer-based models.

Unlike a traditional deep learning model that processes each term separately and outside the context of the sequence, transformers use self-attention to build rich representations of each constituent in the data span, allowing these models to understand the relevance of the location of a term, the relation of one term to the next (even if far away from each other) and more. When trained on larger datasets, these models reach remarkable accuracy and recall for understanding unstructured data like large documents of natural language text.

In the next installation of our Healthcare Fireside Chats, we’re joined by Dell Technologies‘ Global & Federal Healthcare CTO, Michael Giannopoulos. We’ll discuss the rise of unstructured content in healthcare and how Vyasa is applying transformers to improve how the industry can access and gain insight from this valuable data.

Watch more below: