A recent study in The Lancet Oncology showed that AI found 20 percent more cancers on mammogram screens than radiologists alone.
Published in The Lancet Oncology, the study analyzed the scans of 80,000 Swedish women with a moderate hereditary risk of breast cancer who had undergone a mammogram between April 2021 and July 2022. Half of these scans were read by AI and then a radiologist to double-check the findings. The second group of scans was read by two researchers without the help of AI. (Currently, the standard of care across Europe is to have two radiologists analyze a scan before diagnosing a patient with breast cancer.)
The study showed that the AI group detected cancer in 6 out of every 1,000 scans, while the radiologists detected cancer in 5 per 1,000 scans. In other words, AI found 20 percent more cancers than the highly-trained radiologists.
Scientists have been using MRI images (like the ones pictured here) to train artificial intelligence to detect cancers earlier and with more accuracy. Here, MIT's AI system, MIRAI, looks for patterns in a patient's mammograms to detect breast cancer earlier than ever before. news.mit.edu
But even though the AI was better able to pinpoint cancer on an image, it doesn’t mean radiologists will soon be out of a job. Dr. Laura Heacock, a breast radiologist at NYU, said in an interview with CNN that radiologists do much more than simply screening mammograms, and that even well-trained technology can make errors. “These tools work best when paired with highly-trained radiologists who make the final call on your mammogram. Think of it as a tool like a stethoscope for a cardiologist.”
AI is still an emerging technology, but more and more doctors are using them to detect different cancers. For example, researchers at MIT have developed a program called MIRAI, which looks at patterns in patient mammograms across a series of scans and uses an algorithm to model a patient's risk of developing breast cancer over time. The program was "trained" with more than 200,000 breast imaging scans from Massachusetts General Hospital and has been tested on over 100,000 women in different hospitals across the world. According to MIT, MIRAI "has been shown to be more accurate in predicting the risk for developing breast cancer in the short term (over a 3-year period) compared to traditional tools." It has also been able to detect breast cancer up to five years before a patient receives a diagnosis.
The challenges for cancer-detecting AI tools now is not just accuracy. AI tools are also being challenged to perform consistently well across different ages, races, and breast density profiles, particularly given the increased risks that different women face. For example, Black women are 42 percent more likely than white women to die from breast cancer, despite having nearly the same rates of breast cancer as white women. Recently, an FDA-approved AI device for screening breast cancer has come under fire for wrongly detecting cancer in Black patients significantly more often than white patients.
As AI technology improves, radiologists will be able to accurately scan a more diverse set of patients at a larger volume than ever before, potentially saving more lives than ever.
Epigenetic therapeutics could revolutionize medicine in the coming decades. (© kentoh/Adobe)
How exactly does your DNA make you who you are?
It's because of epigenetics that identical twins can actually look different and develop different diseases.
Just as software developers don't write apps out of ones and zeros, the interesting parts of the human genome aren't written merely in As, Ts, Cs and Gs. Yes, these are the fundamental letters that make up our DNA and encode the proteins that make our cells function, but the story doesn't end there.
Our cells possess amazing abilities, like eating invading bacteria or patching over a wound, and these abilities require the coordinated action of hundreds, if not thousands, of proteins. Epigenetics, the study of gene expression, examines how multiple genes work at once to make these biological processes happen.
It's because of epigenetics that identical twins – who possess identical DNA -- can actually look different and develop different diseases. Their environments may influence the expression of their genes in unique ways. For example, a research study in mice found that maternal exposure to a chemical called bisphenol A (BPA) resulted in drastic differences between genetically identical offspring. BPA exposure increased the likelihood that a certain gene was turned on, which led to the birth of yellow mice who were prone to obesity. Their genetically identical siblings who were not exposed to BPA were thinner and born with brown fur.
These three mice are genetically identical. Epigenetic differences, however, result in vastly different phenotypes.
(© 1994 Nature Publishing Group, Duhl, D.)
This famous mouse experiment is just one example of how epigenetics may transform medicine in the coming years. By studying the way genes are turned on and off, and maybe even making those changes ourselves, scientists are beginning to approach diseases like cancer in a completely new way.
With few exceptions, most of the 1 trillion cells that make up your body contain the same DNA instructions as all the others. How does each cell in your body know what it is and what it has to do? One of the answers appears to lie in epigenetic regulation. Just as everyone at a company may have access to all the same files on the office Dropbox, the accountants will put different files on their desktop than the lawyers do.
Our cells prioritize DNA sequences in the same way, even storing entire chromosomes that aren't needed along the wall of the nucleus, while keeping important pieces of DNA in the center, where it is most accessible to be read and used. One of the ways our cells prioritize certain DNA sequences is through methylation, a process that inactivates large regions of genes without editing the underlying "file" itself.
As we learn more about epigenetics, we gain more opportunities to develop therapeutics for a broad range of human conditions, from cancer to metabolic disorders. Though there have not been any clinical applications of epigenetics to immune or metabolic diseases yet, cancer is one of the leading areas, with promising initial successes.
One of the challenges of cancer treatments is that different patients may respond positively or negatively to the same treatment. With knowledge of epigenetics, however, doctors could conduct diagnostic tests to identify a patient's specific epigenetic profile and determine the best treatment for him or her. Already, commercial kits are available that help doctors screen glioma patients for an epigenetic biomarker called MGMT, because patients with this biomarker have shown high rates of success with certain kinds of treatments.
Other epigenetic advances go beyond personalized screening to treatments targeting the mechanism of disease. Some epigenetic drugs turn on genes that help suppress tumors, while others turn on genes that reveal the identity of tumor cells to the immune system, allowing it to attack cancerous cells.
Direct, targeted control of your epigenome could allow doctors to reprogram cancerous or aging cells.
The study of epigenetics has also been fundamental to the field of aging research. The older you get, the more methylation marks your DNA carries, and this has led to the distinction between biological aging, or the state of your cells, and chronological aging, or how old you actually are.
Just as our DNA can get miscopied and accumulate mutations, errors in DNA methylation can lead to so-called "epimutations". One of the big hypotheses in aging research today is that the accumulation of these random epimutations over time is responsible for what we perceive as aging.
Studies thus far have been correlative - looking at several hundred sites of epigenetic modifications in a person's cell, scientists can now roughly discern the age of that person. The next set of advances in the field will come from learning what these epigenetic changes individually do by themselves, and if certain methylations are correlated with cellular aging. General diagnostic terms like "aging" could be replaced with "abnormal methylation at these specific locations," which would also open the door to new therapeutic targets.
Direct, targeted control of your epigenome could allow doctors to reprogram cancerous or aging cells. While this type of genetic surgery is not feasible just yet, current research is bringing that possibility closer. The Cas9 protein of genome-editing CRISPR/Cas9 fame has been fused with epigenome modifying enzymes to target epigenetic modifications to specific DNA sequences.
A therapeutic of this type could theoretically undo a harmful DNA methylation, but would also be competing with the cell's native machinery responsible for controlling this process. One potential approach around this problem involves making beneficial synthetic changes to the epigenome that our cells do not have the capacity to undo.
Also fueling this frontier is a new approach to understanding disease itself. Scientists and doctors are now moving beyond the "one defective gene = one disease" paradigm. Because lots of diseases are caused by multiple genes going haywire, epigenetic therapies could hold the key to new types of treatments by targeting multiple defective genes at once.
Scientists are still discovering which epigenetic modifications are responsible for particular diseases, and engineers are building new tools for epigenome editing. Given the proliferation of work in these fields within the last 10 years, we may see epigenetic therapeutics emerging within the next couple of decades.
