News, Analysis, Trends, Management Innovations for
Clinical Laboratories and Pathology Groups

Hosted by Robert Michel

News, Analysis, Trends, Management Innovations for
Clinical Laboratories and Pathology Groups

Hosted by Robert Michel
Sign In

Researchers Create Artificial Intelligence Tool That Accurately Predicts Outcomes for 14 Types of Cancer

Proof-of-concept study ‘highlights that using AI to integrate different types of clinically informed data to predict disease outcomes is feasible’ researchers say

Artificial intelligence (AI) and machine learning are—in stepwise fashion—making progress in demonstrating value in the world of pathology diagnostics. But human anatomic pathologists are generally required for a prognosis. Now, in a proof-of-concept study, researchers at Brigham and Women’s Hospital in Boston have developed a method that uses AI models to integrate multiple types of data from disparate sources to accurately predict patient outcomes for 14 different types of cancer.

The process also uncovered “the predictive bases of features used to predict patient risk—a property that could be used to uncover new biomarkers,” according to Genetic Engineering and Biotechnology News (GEN).

Should these research findings become clinically viable, anatomic pathologists may gain powerful new AI tools specifically designed to help them predict what type of outcome a cancer patient can expect.

The Brigham scientists published their findings in the journal Cancer Cell, titled, “Pan-cancer Integrative Histology-genomic Analysis via Multimodal Deep Learning.”

Faisal Mahmood, PhD

“Experts analyze many pieces of evidence to predict how well a patient may do. These early examinations become the basis of making decisions about enrolling in a clinical trial or specific treatment regimens,” said Faisal Mahmood, PhD (above) in a Brigham press release. “But that means that this multimodal prediction happens at the level of the expert. We’re trying to address the problem computationally,” he added. Should they be proven clinically-viable through additional studies, these findings could lead to useful tools that help anatomic pathologists and clinical laboratory scientists more accurately predict what type of outcomes cancer patient may experience. (Photo copyright: Harvard.)

AI-based Prognostics in Pathology and Clinical Laboratory Medicine

The team at Brigham constructed their AI model using The Cancer Genome Atlas (TCGA), a publicly available resource which contains data on many types of cancer. They then created a deep learning-based algorithm that examines information from different data sources.

Pathologists traditionally depend on several distinct sources of data, such as pathology images, genomic sequencing, and patient history to diagnose various cancers and help develop prognoses.

For their research, Mahmood and his colleagues trained and validated their AI algorithm on 6,592 H/E (hematoxylin and eosin) whole slide images (WSIs) from 5,720 cancer patients. Molecular profile features, which included mutation status, copy-number variation, and RNA sequencing expression, were also inputted into the model to measure and explain relative risk of cancer death. 

The scientists “evaluated the model’s efficacy by feeding it data sets from 14 cancer types as well as patient histology and genomic data. Results demonstrated that the models yielded more accurate patient outcome predictions than those incorporating only single sources of information,” states a Brigham press release.

“This work sets the stage for larger healthcare AI studies that combine data from multiple sources,” said Faisal Mahmood, PhD, Associate Professor, Division of Computational Pathology, Brigham and Women’s Hospital; and Associate Member, Cancer Program, Broad Institute of MIT and Harvard, in the press release. “In a broader sense, our findings emphasize a need for building computational pathology prognostic models with much larger datasets and downstream clinical trials to establish utility.”

Future Prognostics Based on Multiple Data Sources

The Brigham researchers also generated a research tool they dubbed the Pathology-omics Research Platform for Integrative Survival Estimation (PORPOISE). This tool serves as an interactive platform that can yield prognostic markers detected by the algorithm for thousands of patients across various cancer types.  

The researchers believe their algorithm reveals another role for AI technology in medical care, but that more research is needed before their model can be implemented clinically. Larger data sets will have to be examined and the researchers plan to use more types of patient information, such as radiology scans, family histories, and electronic medical records in future tests of their AI technology.

“Future work will focus on developing more focused prognostic models by curating larger multimodal datasets for individual disease models, adapting models to large independent multimodal test cohorts, and using multimodal deep learning for predicting response and resistance to treatment,” the Cancer Cell paper states.

“As research advances in sequencing technologies, such as single-cell RNA-seq, mass cytometry, and spatial transcriptomics, these technologies continue to mature and gain clinical penetrance, in combination with whole-slide imaging, and our approach to understanding molecular biology will become increasingly spatially resolved and multimodal,” the researchers concluded.  

Anatomic pathologists may find the Brigham and Women’s Hospital research team’s findings intriguing. An AI tool that integrates data from disparate sources, analyzes that information, and provides useful insights, could one day help them provide more accurate cancer prognoses and improve the care of their patients.   

JP Schlingman

Related Information:

AI Integrates Multiple Data Types to Predict Cancer Outcomes

Pan-cancer Integrative Histology-genomic Analysis via Multimodal Deep Learning

New AI Technology Integrates Multiple Data Types to Predict Cancer Outcomes

Artificial Intelligence in Digital Pathology Developments Lean Toward Practical Tools

Florida Hospital Utilizes Machine Learning Artificial Intelligence Platform to Reduce Clinical Variation in Its Healthcare, with Implications for Medical Laboratories

Artificial Intelligence and Computational Pathology

Medical Laboratories and Pathology Groups Outpacing Hospital Administrators on Overcoming the Data Challenges of Precision and Personalized Medicine

Powered by massive data sets, precision medicine is unlocking new insight into the treatment of diseases and conditions. At the same time, pathology groups and medical laboratories are leading the way to creating the storage solutions, analytic tools, and networks that will power tomorrow’s advances in precision medicine

There is exponential growth in the amount of data generated at medical laboratories, pathology groups, and hospital diagnostic facilities. As gene sequencing technologies and tools continue to decrease in price but increase in both speed and accessibility, the volume of data will grow further still.

This has major implications for the field of precision medicine. In order for physicians, hospitals, and clinical laboratories to move forward with precision medicine, to advance, it will be essential that they have sophisticated capabilities in data handling, storage, and analysis.

Yet, hospitals wanting to do more with precision medicine might not be the providers that unravel the technicalities of harnessing this new pool of big data. Pathology groups and medical laboratories are already familiar with the challenges of managing complex data, such as that created by genome sequencing and molecular diagnostic assays. As they continue to bolster their information technology (IT) staffing and infrastructure, these labs are positioned to be the keepers for much of the data driving new medical developments.

Hospital IT Departments Lag in Preparations to Handle Genomic Data

In a Health Catalyst survey, 59% of healthcare executives polled do not believe that precision medicine will play a significant role in their organizations by 2020. Nearly two-thirds of respondents have no plan to integrate genomic data into their electronic health record (EHR) systems.

Despite the trend against genomic data integration, the same survey noted that half of the respondents believe that DNA sequencing could improve patient treatment strategies within their organizations.

However, as medical laboratories have known for some time, obtaining data from molecular and genetic testing is only the beginning. Without the ability to analyze results, communicate information, and store lab test data for access by other parties, the potential benefits of precision medicine will remain unrealized.

Medical Laboratories Leading the Charge for Better Data Handling and Analysis

The good news is that within many health systems, medical laboratories are already adapting IT systems to manage the large data sets required for molecular diagnostics and genetic testing. One of the lessons these labs are learning is that the more molecular and genetic testing the do, the more informatics staff they need.

Simply said, these innovative labs are devoting more physical space to informatics and a larger proportion of the laboratory staff are informatics specialists. In The Dark Report, Gregory J. Tsongalis, PhD, Professor of Pathology and Director of Molecular Pathology at the Theodore Geisel School of Medicine at Dartmouth College, stated, “We just opened a new clinical lab facility of 11,000 square feet at the Geisel School. Over 25% of that space will be devoted to data management. Because of the increasing volume of data generated at this site, there are more computational technologists than lab technicians.” (See The Dark Report, “New Molecular Analyzers to Bring Big Data to All Labs, July 13, 2015.)


Gregory J. Tsongalis, PhD (center right), is Professor of Pathology and Director of Molecular Pathology at the Theodore Geisel School of Medicine at Dartmouth College. (Photo copyright: The Geisel School of Medicine at Dartmouth.)

Gregory J. Tsongalis, PhD (center right), is Professor of Pathology and Director of Molecular Pathology at the Theodore Geisel School of Medicine at Dartmouth College. (Photo copyright: The Geisel School of Medicine at Dartmouth.)

Tsongalis believes that labs themselves are positioned to become the storage providers and gatekeepers for much of the data driving precision medicine. Noting an increasing shift toward staffing informaticists in clinical labs to handle data, he says, “This is probably one aspect of the big data trend where pathologists and lab scientists are ahead of health system administrators.”

Medical Labs Must Address Issues Associate with Handling Genomic Data

Regardless of who holds the increasing amount of genomic data coming out of assays and tests, healthcare providers, laboratories, and vendors must create a set of standards to address both integration with EHR systems, and maintaining privacy and security on a scale never-before addressed.

In an article published in Health Data Management, Deven McGraw, Deputy Director for Health Information Privacy at the Department of Health and Human Services Office for Civil Rights, stressed the importance of data security as the regulatory agency begins work on aspects of the Precision Medicine Initiative. “A robust data security framework will be built in from the start. This is a new model for scientific research, but it is not widespread,” McGraw stated.

The Action Collaborative on Developing Guiding Principles for Integrating Genomic Information into the Electronic Health Record Ecosystem also is piloting a collaborative between Intermountain Healthcare, ARUP Laboratories, and Cerner, to examine privacy standards as well as standards for integration and representation of genomic data sets.

As pools of healthcare data and medical laboratory test results continue to grow and clinical laboratories continue to explore new methods of genetic testing and analysis, partnerships such as the Action Collaborative will remain critical opportunities for medical laboratories and pathology groups to use their current advantage to further their future roles in precision medicine.

—Jon Stone

Related Information:

Survey: Most Healthcare Organizations Unprepared for Precision Medicine 

Big IT Challenges Ahead for Precision Medicine 

About the Precision Medicine Initiative Cohort Program 

Fact Sheet: President Obama’s Precision Medicine Initiative 

DIGITizE: Displaying and Integrating Genetic Information Through the EHR 

Newer, Smaller Analyzers Will Bring Big Data to Labs