Shape shifting: Researchers at MIT use a technique called dynamic time warping to compare the shape of individual waves in an electrocardiogram, shown here. This approach aligns waves based on features that correspond to the same underlying physiological activity, allowing an automated and accurate comparison. The measure, dubbed morphological variability, appears to predict who is at greatest risk of death after a heart attack.   Credit: Collin Stultz


EKG patterns show who will likely die after a heart attack.


MIT Technology Review, December 16, 2009, by Emily Singer  —  A new approach to analyzing electrocardiograms–a ubiquitous test of the heart’s electrical function–could predict who is most likely to die after a heart attack. Researchers at MIT found that measuring how much the shape of the electrical waveform varies from beat to beat identifies high-risk patients better than existing risk factors. If the findings hold up in further clinical trials, the technology could be used to figure out which heart attack patients need the most aggressive treatment.

Scientists hope the same approach will eventually help them predict when a healthy person is likely to suffer a cardiac problem. They are working with Texas Instruments to integrate the software into the new generation of wearable heart monitors.

The research also shows how computational analysis can glean useful information from the reams of medical data routinely collected and ignored. “It’s a very novel approach,” says Jean-Philippe Couderc, biomedical engineer at the University of Rochester, who was not involved in the project. “It’s a unique way of looking at how the electrocardiogram varies on a beat to beat basis.”

Electrocardiograms record the heart’s electrical activity through sensors placed on the chest. Cardiologists can then spot abnormal heart rhythms by visually inspecting the resulting waveform for major features linked to the function of the top and bottom chambers of the heart, as well as the heart’s ability to “reset” itself between beats. While some simple algorithms exist to analyze this data, they are notoriously inaccurate. “Cardiologists routinely ignore them,” says Collin Stultz, a professor at MIT, as well as a practicing cardiologist, who is involved in the project.

To determine if more subtle features within electrocardiogram data could provide useful clinical information, Stultz, John Guttag, also at MIT, and Zeeshan Syed, now at the University of Michigan, started with a large data set of 24-hour electrocardiogram recordings collected at Brigham and Women’s Hospital in Boston as part of a clinical trial for a new drug. Employing a number of computational techniques, including signal processing, data mining, and machine learning, the researchers developed a way to analyze how the shape of the electrical waveform varies, a measure they dubbed morphological variability. At the heart of the approach is a method called dynamic time warping, used in speech recognition and more recently in genome analysis, which allows researchers to align and compare individual beats. “We compute the differences for every pair of beats,” says Stultz. “If there is lots of variability, that patient is in bad shape.”

The team then applied the algorithm they had developed to a second set of electrocardiogram recordings and found that patients with the highest morphological variability were six to eight times more likely to die after a heart attack than those with low variability. “We found that it consistently works as well or better than commonly accepted risk metrics that physicians use,” says Stultz, including diabetes, age and smoking status, as well as cardiac ultrasound and various blood tests.

Researchers are now planning a number of new clinical studies, including assessing morphological variability in healthy people to get a sense of how this measure varies among a normal population. (The existing research was done on recordings from patients with a history of heart disease.) In addition, the team is modifying the algorithm to shorten the amount of data needed to make the predictions from about 10 hours’ worth of EKG data to less than one. “We hope to get it down to half an hour, which is within the realm of a doctor’s visit,” says Guttag.

They also want to determine whether measuring morphological variability in patients at high risk of developing cardiovascular disease can predict risk of heart attack or death in those who have not yet had a cardiac event.

“It’s an extra test, but it’s one that is cheap, because the data is already available and the analysis can be performed on existing computers,” says Guttag. But the technology still faces a number of hurdles. A prospective trial is needed to confirm that morphological variability is a clinically useful marker. And researchers need to figure out how cardiologists can easily access and analyze the information–the different devices currently used to record EKGs typically lock the data behind protective software.

In the longer term, the researchers aim to incorporate this type of analysis into a new generation of heart monitors currently under development. These monitors are intended to be so small, cheap, and easy to use that people wear them all the time–similar to the heart-rate monitors athletes often wear but able to collect more sophisticated information.

The team is collaborating with Texas Instruments to develop a wearable monitor with an embedded chip to calculate morphological variability in real time. Such a device might be used to signal a patient to take additional medication or signal an implanted device to deliver a jolt of electricity to the heart.

Cardiologists themselves may prove to be a serious obstacle in bringing this type of analysis to clinical practice. Not only is morphological variability too subtle a measure to detect visually, it is not yet clear what it represents physically in the heart. “It’s going to be difficult for a nonengineer to link this measurement to specific abnormal electrophysiological phenomena,” says Couderc. “That needs to be clarified and communicated to the medical field.”


Making Medicine Personal

MIT Technology review, 12/16/09

A number of scientists bared their genetic souls recently as part of the Personal Genome Project, a study at Harvard University Medical School. They were among the first of the eventually 100,000 volunteers who will agree to place their genetic profiles on the Internet. 

Genetic profiling can provide information on what diseases may befall us. And knowledge of an individual’s genetic makeup may also help scientists figure out how to treat diseases-part of an emerging field known as personalized medicine.

As many doctors freely admit, says Julie Johnson, director of the Center of Phamacogenomics at the University of Florida (UF), prescribing medicine is “more of an art than a science.” Approved drugs work-but not 100 percent of the time, and not for 100 percent of the population. Some people have no response to certain drugs, and others experience severe side effects.

What determines whether a particular treatment is effective or leads to severe side effects is our genes, scientists believe. Personalized medicine holds the promise of tailored medical treatments based on genetic information, rather than a one-size-fits-all approach.

The UF center participated in studies on warfarin, a blood thinner prescribed for millions of Americans to prevent heart attack or clotting after a heart attack. Too little of the drug causes a risk of clotting, and too much can cause excessive bleeding. “There’s a very narrow window, and there’s a great deal of variability among patients,” says Johnson. “A lot of work in the past decade has uncovered several genes that help explain a great deal of that variability.” In 2007, the FDA cleared a genetic test for sensitivity to warfarin to help doctors prescribe the correct dosage, although the tests are not yet widely implemented.

The UF center is also focusing research on drugs prescribed for hypertension, in an attempt to find the genes that “will predict how much a person’s blood pressure will go down if they’re administered certain medicines,” says Johnson.

Speeding the Process

Part of what has contributed to the increasing interest in personalized medicine is the speed and cost of sequencing genomes. The first human genome took many years and millions of dollars to sequence. The price has already dropped into the thousands instead of millions of dollars, and it’s expected to continue to fall. The journal Science listed “faster, cheaper genome sequencing” as one of the top scientific advances in 2008.

These advances have increased the speed of research in the field. John Reed, the president and CEO of Burnham Institute for Medical Research, a center with campuses in San Diego, CA, and Orlando, FL, says that the Florida campus has engaged in major initiatives related to personalized medicine. While Burnham’s research has traditionally focused on cancer and on neurodegenerative and inflammatory diseases, the scientific team is expanding into obesity, diabetes, and metabolism research.

“We all have friends who can eat french fries every day and never gain weight, while the rest of us will have a hard time getting the belt to fit,” says Reed. “There are genetic differences in how we metabolize food-individual metabolic rates, hormone signaling-that’s all just being worked out.” Burnham is partnering with the clinical research institute at Florida Hospital, particularly the diabetes center, to engage in research on the metabolic systems of the patients there.

A related field of research involves investigating which chemicals can affect the actions of proteins, encoded by specific genes. This is a natural path to drug discovery, but it can also aid in genomic research. “A chemical probe can be used in basic research to help identify the role of a protein or a pathway, aiding in understanding the biology of a particular gene,” says Patrick Griffin, chair of Molecular Therapeutics at Scripps Florida, a campus of Scripps Research Institute headquartered in California.

The National Institutes of Health (NIH) funds four molecule-screening centers in the United States to rapidly test a library of chemicals against specific proteins. Scripps Florida operates one of the four centers.

Burnham operates a second of those NIH molecule-screening centers at both its California and Florida research centers. Currently, its screening output can tackle half a million chemicals in one day, but the new system being developed in Orlando will be able to handle as many as 2.2 million chemicals a day.

The fields of genome research and rapid drug discovery are coming together to enhance each other, says Reed. “We’ll be able to, with far more accuracy, define for whom a drug is really going to work, and to avoid a lot of trial and error that we experience when we’re confronted with a health issue.” He and other researchers in the field see a time not too far in the future when understanding individual genomes will lead to better, more effective medical treatments for everyone.

Broad Foundation gift will help guide new research toward clinical applications


Pasadena and Westwood, Calif.-The California Institute of Technology (Caltech) and the University of California, Los Angeles (UCLA) have announced the establishment of the Joint Center for Translational Medicine (JCTM), which will advance experimental research into clinical applications, including the diagnosis and therapy of diseases such as cancer.

Initial funding for the new center comes from a two-year, $5 million gift from The Eli and Edythe Broad Foundation.

“The strengths of both institutions will be brought together in this new center to help move discoveries from research into clinical practice,” says David Baltimore, Nobel laureate and the Robert Andrews Millikan Professor of Biology at Caltech. Baltimore will be the center’s director.

Owen Witte, a Howard Hughes Medical Institute investigator and director of the Broad Stem Cell Research Center at UCLA, will serve as deputy director of the new center. The center will build upon UCLA’s strength and international reputation in conducting translational research, including development of the molecularly targeted drugs Herceptin, Gleevec, Avastin, and Sprycel. The program, Witte says, will take the best science from the laboratories at Caltech and UCLA and will transform it into new and more effective therapies for debilitating diseases.

“The move to combine the expertise and experience at these two premier research institutions will set the standard for the Los Angeles area,” he adds. “This new center is the natural evolution of several research collaborations between UCLA and Caltech and will result, we hope, in many new options for people with a host of diseases.”

The first project of the center will be the investigation into a potentially revolutionary treatment for late-stage melanoma, in which the body’s killer immune cells are programmed to recognize and destroy tumor cells. This work started in 2006 as joint research between Caltech and UCLA and led to the idea for the center. The melanoma research has already advanced to ongoing clinical trials involving a half-dozen patients.

“We saw the success of the melanoma research program and asked ourselves, ‘Is there something else we can do for other diseases?'” says Baltimore. “The Broad Foundation has always looked for programs that elevate the quality of research in Los Angeles, and this new center will go a long way toward enhancing the region’s reputation for medical research.”

“We have a great deal of admiration and respect for Dr. Baltimore and Dr. Witte, and this new center brings together some of the brightest minds in science and medicine,” says Eli Broad, founder of The Eli and Edythe Broad Foundation and a major funder of the Broad Center for the Biological Sciences at Caltech, the Eli and Edythe Broad Center of Regenerative Medicine and Stem Cell Research at UCLA, the Eli and Edythe Broad Center for Regenerative Medicine and Stem Cell Research at USC, the Eli and Edythe Broad Center of Regeneration Medicine and Stem Cell Research at the University of California, San Francisco, and the Broad Institute for Biomedical Research in partnership with Harvard and MIT.

In addition to Baltimore’s research, the work of Caltech’s Mark Davis, the Warren and Katharine Schlinger Professor of Chemical Engineering, and James Heath, the Elizabeth W. Gilloon Professor and professor of chemistry, is already generating joint translational opportunities with UCLA. Researchers at UCLA involved in joint efforts with the three Caltech labs include Antoni Ribas, associate professor of hematology/oncology; Caius Radu, assistant professor of molecular and medical pharmacology; and Michael Phelps, the Norton Simon Professor, chair of the Department of Molecular and Medical Pharmacology, and inventor of positron emission tomography scanning.

“The hope is to bring in additional researchers who will use the center as a vehicle for moving their work from a research phase to one with clinical potential,” says Baltimore.

To help foster that process, the center will immediately begin a Translational Acceleration Grants (TAG) program, which will offer seed grants for work by Caltech and UCLA faculty members. The awards will support highly competitive applications proposing research that has the potential to move toward clinical applications. Each TAG will consist of an initial one-year award of up to $100,000 to support direct research costs, with the possibility of a one-year extension.

An advisory board will assist in the grant review process as well as policy decisions for the new center. Six members, three from each institution, sit on the board. The Caltech board members are Peter Dervan, the Bren Professor of Chemistry; Ray Deshaies, professor of biology; and Scott Fraser, the Anna L. Rosen Professor of Biology and professor of bioengineering, and director of the Donna and Benjamin M. Rosen Bioengineering Center. The UCLA board members are Judith C. Gasson, director of UCLA’s Jonsson Comprehensive Cancer Center; Donald Kohn, director of the Gene Medicine Program; and Bruce Dunn, a professor in the Department of Materials Science and Engineering.

“We expect this new center to strengthen our current work on both campuses and identify future translational work,” says Baltimore, “as well as create stronger ties between Caltech and UCLA.”

Public release date: 16-Dec-2009

Contact: Jon Weiner
              626-395-3226         626-395-3226
California Institute of Technology

Kim Irwin
              310-206-2805         310-206-2805
University of California – Los Angeles