AI-Powered Diagnostic Tool Predicts Pancreatic Cancer Up to 3 Years in Advance

Given how difficult pancreatic cancer can be to diagnose, the tool could be life-saving.

AI-Powered Diagnostic Tool Predicts Pancreatic Cancer Up to 3 Years in Advance

By Adrianna Nine  in Extremetech

Pancreatic cancer might be relatively rare, but it’s one of the world’s leading causes of cancer-related death. Thanks to the organ’s placement deep within the abdomen, tumors on the pancreas can be difficult to detect, forcing doctors to rely on expensive and invasive imaging or blood tests instead. If these strategies are implemented too late, a patient’s odds of successfully completing treatment are slim.

A new tool powered by artificial intelligence might make early detection easier and more accessible. Researchers from Harvard Medical School and the University of Copenhagen partnered with the VA Boston Healthcare System and the Dana-Farber Cancer Institute to build a screening program that scans people’s medical records to determine their likelihood of developing pancreatic cancer. According to a paper published Tuesday in Nature Medicine, the tool can predict future pancreatic cancer diagnosis up to three years ahead with 88% accuracy.

The researchers trained an AI model on roughly 9 million patients’ health records obtained from the Danish National Patient Registry (DNPR) and the US Veterans Affairs database (US-VA). Both of these databases were used to train the model separately. Of the 6 million patients in the DNPR, 23,985 had been diagnosed with pancreatic cancer; of the 3 million US-VA patients, 3,900 had received the same diagnosis. Using this data, the AI model was trained to pick up on disease codes, comorbidities, and timelines that might indicate a future pancreatic cancer diagnosis—even if the disease codes themselves had nothing to do with the pancreas. ... ' 

A New gel-based treatment for Glioblastoma

New Therapy Found to Prevent Aggressive Brain Cancer Recurrence in Mice

A new gel-based treatment for glioblastoma—a particularly lethal form of brain cancer—has shown promise in the lab.

By Adrianna Nine April 27, 2023

CT brain scan showing a glioblastoma tumor. 

Glioblastoma mass shown on a CT scan. Credit: Dr. P Marazzi/Science Photo Library/Getty Images

A new gel-based treatment for glioblastoma—a highly aggressive form of brain cancer—has shown to be 100% effective at preventing recurrence in mice. Researchers hope the therapy will translate well into human physiology, where it could help resolve tens of thousands of cancer diagnoses every year.

Glioblastoma manifests as a tumor growing on the brain or spinal cord. While many glioblastoma patients have the tumor surgically removed, the mass often returns, even in cases involving post-surgical radiation or chemotherapy. The disease is so persistent that the average patient lives only 12 to 16 months after diagnosis, making glioblastoma one of the most lethal forms of cancer currently understood.

Researchers at Johns Hopkins University are working to improve patients’ life expectancies using an injectable gel that blocks cancer’s path. According to a paper published Tuesday in Proceedings of the National Academy of Sciences, the gel is made up of nano-sized filaments derived from the drug paclitaxel, which is used alongside chemotherapy to treat other forms of cancer. The gel serves as a vehicle for aCD47, an antibody that prompts macrophages to ingest tumor cells.   ... '

Video Game Developers and Cancer Researchers Team Up

 Even seems like a natural collaboration these days.    Had been wondering how much of that is now common between game players/designers and science in general.

Cambridge University Researchers Develop VR Tool for Cancer Treatment

Video game developers and cancer researchers have teamed up to turn spreadsheet data into highly detailed VR imagery of cancer cells.      By Marco Marcelline

Virtual reality software has become an unlikely tool in the fight against cancer. 

In a bid to help doctors better understand how to treat cancer, video game designers and cancer researchers have teamed up at the University of Cambridge, England, to turn spreadsheet data into highly detailed VR imagery of cancer cells, ITV reports(Opens in a new window). 

The university’s IMAXT Laboratory has transformed brain-crunching numbers and data into an interactive 3D picture of a tumor that makes it easy for researchers to differentiate between cancer cells, as each type of cell is colored or shaped differently. 

With the help of a VR headset, doctors and researchers can essentially step inside patients’ tumors, making it easier to assess the severity and origin of the cancer cells. The aim of the tool, its makers say, is to give a better insight into how tumors can be treated.  ... ' 

Cambridge Researchers Develop VR Tool for Cancer Treatment

Appears to be of use  ..

Cambridge Researchers Develop VR Tool for Cancer Treatment

PC Magazine

Marco Marcelline, December 29, 2022

Video game designers and cancer researchers have teamed up at the U.K.'s University of Cambridge IMAXT Laboratory to turn spreadsheet data into highly detailed virtual reality (VR) imagery of cancer cells. With a VR headset, users can essentially step inside patients’ tumors, making it easier to assess the severity and origin of the cancer cells and glean better insight into how tumors can be treated. ..  '

Tool Uncovers Cancers

 Tool Uncovers Cancer-Driving Structural Variations

Weill Cornell Medicine Newsroom, September 26, 2022

Weill Cornell Medicine researchers created the CSVDriver software to identify cancer-generating structural variants (SVs) from tumor samples via DNA sequence analysis. The software maps and analyzes SV locations in tumor DNA datasets; the researchers applied CSVDriver to a dataset of 2,382 genomes from 32 different cancer types, analyzing the cancer genomes from different organ systems independently. The outcomes verified the likely cancer-producing roles of 47 genes, and suggested 26 other genes as likely cancer drivers. "The general idea here was to model the distribution of background mutations that we would expect for a given cancer type, and then identify, as candidate driver locations, regions where mutations occur more often than expected in a large fraction of patients," said Weill Cornell Medicine's Alexander Martinez-Fundichely.

Personalized Cancer Treatment

Is this the future?   

Personalizing cancer treatment with quantum computing

Press Release / August 10, 2021

As a partner of the Fraunhofer Competence Network Quantum Computing, the German Cancer Research Center (DKFZ) is planning to use the quantum computer in Ehningen, Baden-Württemberg, to develop individually effective cancer treatment methods in the future.

Cancer patients’ medical records can often comprise up to 100 terabytes of individual — and usually very heterogeneous — data, including blood and tumor values, personal indicators, sequencing and treatment data, and much more besides. Up to now, it has been virtually impossible to use this wealth of information efficiently due to a lack of appropriate processing mechanisms. As a result, the possibility of using promising personalized treatment approaches remains purely theoretical for many cancers, and patients are still receiving standard treatments.

Now, the German Cancer Research Center (DKFZ) in Heidelberg is planning to use quantum computing to drive forward research in this area: “We want to explore how we can systematically process and use this heterogeneous data with the aid of a quantum computer, so that we can identify new and more targeted options for patients who do not respond so well to immunotherapies. Ultimately, we are asking which patient can benefit from which treatment and how,” says Dr. Niels Halama, Head of Department of Translational Immunotherapy at the German Cancer Research Center (DKFZ) and Senior Physician at the German National Center for Tumor Diseases. Linked to this topic are some applied research questions: Which signaling cascades and biological processes play a role in the disease? How can we use these to select a treatment on an individual basis? What kinds of problems actually lend themselves to being solved by quantum computers?

From a simulator to a real quantum computer

The DKFZ team has already worked out the mathematical principles and carried out some initial work using other globally available systems and simulators. According to Halama, however, there is a huge difference between working on a simulator with perfect qubits and working on a real quantum computer such as IBM Q System One in Ehningen. It is only with the latter that you can see how stable things are at a certain level of complexity, where the pitfalls are and what is possible.  ... ' 

AI for Personalized Cancer Vaccines

 Seems quite a big move forward.    Will we be able to learn which amino acid sequences will work best to fight cancer? 

Using Machine Learning to Develop Personalized Cancer Vaccines

University of Waterloo Cheriton School of Computer Science (Canada)  via ACM

Researchers at Canada's University of Waterloo Cheriton School of Computer Science are applying machine learning to identify tumor-specific neoantigens, which could lead to personalized cancer vaccines. Cheriton's Hieu Tran said the team used a model similar to natural language processing to ascertain neoantigens' amino acid sequences based on one-letter amino acid codes. The researchers used the DeepNovo recurrent neural network to predict amino acid sequences, which Tran said expanded the predicted immunopeptidomes of five melanoma patients by 5% to 15%, based solely on data from mass spectrometry—and personalized the neoantigens to each patient.

Detecting Cancer Cells

A Computational means of differentiating cells.

Computational Tool Reliably Differentiates Between Cancer, Normal Cells From Single-Cell RNA-Sequencing Data     By The University of Texas MD Anderson Cancer Center, January 28, 2021

Researchers at The University of Texas MD Anderson Cancer Center have developed a computational tool to reliably distinguish between cancerous and normal cells in tumor samples. Scientists can use the CopyKAT (copy number karyotyping of aneuploid tumors) tool to more easily analyze gene-expression data from large single-cell RNA-sequencing experiments.

MD Anderson's Nicholas Navin noted that CopyKAT mines the data to uncover abnormal chromosome numbers typical of most cancers, and identifies distinct subpopulations, or clones, within the cancer cells. Said Navin, "By applying this tool to several datasets, we showed that we could unambiguously identify, with about 99% accuracy, tumor cells versus the other immune or stromal cells present in a mixed tumor sample."

Added former MD Anderson researcher Ruli Gao, "We hope this tool will be useful to the research community to make the most of their single-cell RNA-sequencing data and to drive new discoveries in cancer."

From The University of Texas MD Anderson Cancer Center

IBM Gives Cancer AI to Open Source

Like the potential of clearly beneficial approaches being shared this way.    What outcomes have come out of this work to date?

IBM Gives Cancer-Killing Drug AI Project to the Open Source Community 
ZDNet
Charlie Osborne

IBM has released to the open source community three artificial intelligence (AI) projects designed to address the challenge of curing cancer. The projects, led by researchers at IBM's Computational Systems Biology Group in Switzerland, involve developing AI and machine learning approaches to help accelerate the understanding of the leading drivers and molecular mechanisms of different cancers. The first project, PaccMann, is working to develop an algorithm that can automatically analyze chemical compounds and predict which are most likely to overcome cancer strains. The second project, "Interaction Network infErence from vectoR representATions of words" (INtERAcT), aims to develop a tool that can automatically extract information from the thousands of papers published every year on cancer research. The third project, "pathway-induced multiple kernel learning," focuses on an algorithm that uses datasets describing what is currently known about molecular interactions to predict the prognosis of cancer patients.  .... "

Cancer Algorithm and Game Theory

New heath care approaches.

Cancer algorithm uses game theory to Double Survival time

Using algorithms to monitor cancer evolution and apply game theory to their treatment has doubled the survival time of men with advanced prostate cancer    By Andy Coghlan in NewScientist

Approaching cancer treatment as a game has doubled the survival time of men with advanced prostate cancer. This achievement could mark the start of using game theory to target a range of cancers more cleverly.

“This approach is elegant and exciting, and shows real promise to delay treatment failure,” says Charles Swanton at the Francis Crick Institute in London.

People with cancer aren’t usually killed by their initial tumour, but by the rapidly evolving secondary tumours that occur once the disease …  "

Cancer Research with Deep Learning

In the CACM:

" .... Oak Ridge National Laboratory (ORNL) researchers are applying deep-learning techniques to automate how information is collected from cancer pathology reports documented across a nationwide network of cancer registry programs.

Georgia Tourassi, director of the Health Data Sciences Institute at ORNL, led a team focused on software that can identify valuable information in cancer reports faster than manual methods.

The machine-learning technique leverages algorithms, big data, and the processing power of the Titan supercomputer at the Oak Ridge Leadership Computing Facility. Using a dataset of nearly 2,000 pathology reports, researchers trained a deep-learning algorithm to simultaneously carry out two closely related tasks. .... " 

60 Minutes on AI

Last nights's 60 Minutes spent about thirty minutes on AI.  Watson got a good chunk of the coverage with Cancer diagnosis.  Also Carnegie Mellon being Innovative.    As usual, was quite thin in its presentation.  Quite a few mentions of what was being done, but hardly any 'How'.  Questioner came from an uninformed place. Bare mention of cautions of AI being dangerous.   Also barely mentioning taking over jobs.  Mostly useful to see how this is being presented to decision makers you are talking to.

On Patient Behavior

In every business where there are customers, its about their behavior.   Even in healthcare. Knowledge@Wharton talks about a new book:  How Patients Think: A Science-Based Strategy for Patient Engagement and Population Health. by Andrea LaFountain

Knowledge@Wharton: In this age of technology and big data, why is there still that disconnect between healthcare providers and patients?

" .... Andrea LaFountain: You kind of answered the question. Big data can be great, but it can also be an enemy depending on how we use the data. There’s an overreliance on what we call “claims data.” We look at the activities of the patient. We can see whether they are visiting the doctor or filling prescriptions, but it doesn’t give us the explanation behind the activities that we’re seeing. That’s what’s missing.

The data can show us the results of their decision. It doesn’t tell us why they are making the decision. So we can look at the data, and we can see that 26% of women with breast cancer stop taking their treatment early.

We can see that very clearly in the data. But how are they making that decision? That is not available in the data. So we speculate and we kind of impose our reasons why we think they are making that decision without actually getting into the scientific reasons behind why they are making those decisions.  .. " 

Data World Provides Data Resources

Activity in the semantic data world,   Dean Allemang mentioned and quoted.   Toward an improvement of the use of data assets.    An indication of increased open ontology development?

PRESS RELEASE
Austin, TX – July 11, 2016 – data.world, a new company that makes it fast and easy to find and use data to solve problems, has officially launched today. Led by Bazaarvoice founder (NASDAQ: BV) and former CEO Brett Hurt and three former HomeAway executives, data.world is building the most meaningful, collaborative, and abundant data resource in the world by dismantling the barriers between people and data. With data.world, people who work with data can quickly and easily discover, prepare, and share it, as well as effectively collaborate in real time to solve important problems.   .... 

“Open data, which is data that can be freely used, modified, and shared by anyone for any purpose, is shaping up to be one of the most important forces impacting humanity today,” said Hurt, co-founder and CEO of data.world. “This movement can speed our cure for cancer, help keep governments accountable, curtail climate change, and positively impact other important world issues. We’ve built data.world to facilitate these breakthroughs – by allowing professional and amateur data scientists, analysts, and researchers to instantly find, use, and share data.”   .... 

“data.world is tackling a fundamental problem: how to make data interoperable to help people and machines unlock data’s value faster. This is a significant step forward for the semantic web community – and a more comprehensive approach than I’ve seen before.” – Dean Allemang, semantic web expert, author and Principal Consultant at Working Ontologist, LLC.    ... " 

The Patient Will see You Now

In MIT Sloan:

Digital technology empowers patients to set their own course of care.

" .... Digital technology is empowering patients to participate in developing their own treatment plans, but only if the organization’s culture is ready and willing, says Kristin Darby, CIO of Cancer Treatment Centers of America. “We crave constructive disruption, so we are always challenging ourselves with the question, ‘how can technology positively impact our patients?’ If there’s value for the patient, we’re interested and we dig deeper.” Darby spoke with MIT Sloan Management Review guest editor Gerald C. Kane about how digital disruption has changed her company's leadership perspective. ... "

A Crusade Against Multiple Regression Analysis

In the Edge
Yes, well known, but often ignored,  is all the context in the model?  Again in the realm of misusing statistics.  A lengthy conversation with Richard Nisbett.  A crusade, he says, that may be worth taking note of.

" .... The thing I’m most interested in right now has become a kind of crusade against correlational statistical analysis—in particular, what’s called multiple regression analysis. Say you want to find out whether taking Vitamin E is associated with lower prostate cancer risk. You look at the correlational evidence and indeed it turns out that men who take Vitamin E have lower risk for prostate cancer. Then someone says, "Well, let’s see if we do the actual experiment, what happens." And what happens when you do the experiment is that Vitamin E contributes to the likelihood of prostate cancer. How could there be differences? These happen a lot. The correlational—the observational—evidence tells you one thing, the experimental evidence tells you something completely different. ... " 

Google Developing Pill to Detect Disease

In Wired.  An idea that was first posed back in the earlier AI days.  Is it now close to happening?

" ... Google is attempting to develop a pill that would send microscopic particles into the bloodstream in an effort to identify cancers, imminent heart attacks, and other diseases.

Andrew Conrad, the head of life sciences inside the company’s Google X research lab, revealed the project on Tuesday morning at a conference here in Southern California. According to Conrad, the company is fashioning nanoparticles—particles about one billionth of a meter in width—that combine a magnetic material with antibodies or proteins that can attach to and detect other molecules inside the body. The idea is that patients will swallow a pill that contains these particles, and after they enter the bloodstream—attempting to identify molecules that would indicate certain health problems—a wearable device could use their magnetic cores to gather them back together and read what they’ve found. ... " 

Autodesk Prints 3D Virus

Did not know that Autodesk had a genetic engineer.  Worked with them in the area of displaying marketing data for the enterprise.  From 3D Print: " ... Weaving his way through a virtually invisible and widely unknown world of microscopic cells and futuristic technology, Autodesk’s genetic engineer Andrew Hessel is working to fight cancer on a new level: a personalized, affordable one, using 3D printed oncolytic viruses, which literally break cancer cells apart.

He makes fighting cancer sound very simple with the idea of viral engineering, which encompasses using software to design and make viruses — and 3D print them. Obviously, there is quite a body of research and work that goes into making that a reality. ... "

Developing Analytic Talent

Just received.  A quick look indicated that this is a useful, only mildly technical look at what is being called Data Science.  A guide and handbook for anyone interested into the topic. Will review further and post that here soon.

Developing Analytic Talent: Becoming a Data Scientist
by Vincent Granville

Their decription:  " ... Harvard Business Review calls it the sexiest tech job of the 21st century. Data scientists are in demand, and this unique book shows you exactly what employers want and the skill set that separates the quality data scientist from other talented IT professionals. Data science involves extracting, creating, and processing data to turn it into business value. This guide discusses the essential skills, such as statistics and visualization techniques, and covers everything from analytical recipes and data science tricks to common job interview questions, sample resumes, and source code.

The applications are endless and varied: automatically detecting spam and plagiarism, optimizing bid prices in keyword advertising, identifying new molecules to fight cancer, assessing the risk of meteorite impact. Complete with case studies, this book is a must, whether you're looking to become a data scientist or to hire one. ... " 

Watson and Brain Cancer Assistance

In IEEE Spectrum:  Watson: takes on the genetics of brain cancer.  The use of an 'AI' as an assistant is a good model, which we used often for related models.  The challenge will be to get professionals to utilize the capability usefully.  Human expertise molding the combination of data and analytics.  " ... Twenty patients with an aggressive form of brain cancer will have a new doctor on their medical team: the learned geneticist known as IBM Watson. In a collaboration announced today between IBM and the New York Genome Center, IBM's Jeopardy-beating AI will analyze the genomes of those 20 patients in hopes of providing insights for their oncologists.  ... "