cancer diagnosis – NIH Director's Blog (original) (raw)
AI Model Takes New Approach to Performing Diagnostic Tasks in Multiple Cancer Types
Posted on October 3rd, 2024 by Dr. Monica M. Bertagnolli
Credit: Donny Bliss/NIH, Adobe Stock
In recent years, medical researchers have been looking for ways to use artificial intelligence (AI) technology for diagnosing cancer. So far, most AI models have been developed to perform specific tasks in cancer diagnosis, such as detecting cancer presence or predicting a tumor’s genetic profile in certain cancer types. But what if an AI system could be more flexible, like a large language model such as ChatGPT, performing a variety of diagnostic tasks across multiple cancer types?
As reported in the journal Nature, researchers have developed an AI system that can perform a wide range of cancer evaluation tasks and outperforms current AI methods in tasks like cancer cell detection and tumor origin identification. It was tested on 19 cancer types, leading the researchers to refer to it as “ChatGPT-like” in its flexibility. According to the research team, whose work is supported in part by NIH, this is also the first AI model based on analyzing slide images to not only accurately predict if a cancer is likely to respond to treatment, but also to validate these predictions across multiple patient groups around the world.
Today, when doctors order a biopsy to find out if cancer is present, those samples are sent to a pathologist, who examines the tissues or cells under a microscope to determine if they are cancerous. The team behind this AI model, led by Kun-Hsing Yu, Harvard Medical School, Boston, recognized that pathologists must analyze a wide variety of disease samples. To make accurate diagnoses in different cancer types, they must take many subtle factors into account.
Most earlier attempts to devise an AI model to analyze tissue samples have depended on training computers to recognize one cancer type at a time. In the new work, the researchers developed a more general-purpose pathology AI system that could analyze a broader range of tissues and sample types. To develop their Clinical Histopathology Imaging Evaluation Foundation (CHIEF) model, the researchers used an AI approach known as self-supervised learning. In this method, a computer is given large volumes of data, in this case 15 million pathology images, to allow it to identify intrinsic patterns and structures. This process allows a computer to “learn” from experience to identify informative features in a vast data set.
The tool was then trained further on more than 60,500 whole-slide images in tissues collected from 19 different parts of the body—such as the lungs, breast, prostate, kidney, brain, and bladder—to bolster the model’s ability to capture similarities and differences among cancer types. This training data was in part comprised of data from The Cancer Genome Atlas (TCGA) program and the Genotype-Tissue Expression (GTEx) Project, both NIH-supported resources. The researchers directed the model to consider both the image as a whole and its finer details, enabling it to interpret the image in a broader context than one region. They then put CHIEF to the test, using another 19,491 whole-slide images from 32 independent slide sets collected from 24 hospitals around the world.
They found that CHIEF worked equally well no matter how the samples were collected (biopsy or surgical excision) and in different clinical settings. In addition to detecting cancers and predicting a cancer’s tissue of origin, CHIEF also predicted with 70% accuracy whether a tissue carried one among dozens of genetic mutations that are commonly seen in cancers. CHIEF showed an ability to predict whether a sample contained mutated copies of 18 genes that oncologists use to make treatment decisions. CHIEF could predict better than earlier models how long a patient was likely to survive following a cancer diagnosis and how aggressively a particular cancer would grow.
This is all good news, but there’s much more work ahead before an AI model like this could be used in the clinic. Next steps for the researchers include training the model on images of tissues from rare cancers, as well as from pre-cancerous and non-cancerous conditions. With continued development and validation, the researchers aim to enable the system to identify cancers most likely to benefit from targeted or experimental therapies in hopes of improving outcomes for more people with cancer in diverse clinical settings around the world.
Reference:
Wang X, et al. A pathology foundation model for cancer diagnosis and prognosis prediction. Nature. DOI: 10.1038/s41586-024-07894-z (2024).
NIH Support: National Institute of General Medical Sciences, National Cancer Institute
Posted In: Health, Science, technology
Tags: AI, artificial intelligence, cancer, cancer diagnosis, computer learning, genetic mutations, genetics, imaging, pathology, technology
Insurance Status Helps Explain Racial Disparities in Cancer Diagnosis
Posted on January 21st, 2020 by Dr. Francis Collins
Credit: iStock/jmangostock
Women have the best odds of surviving breast cancer if their disease is caught at an early stage, when treatments are most likely to succeed. Major strides have been made in the early detection of breast cancer in recent years. But not all populations have benefited equally, with racial and ethnic minorities still more likely to be diagnosed with later-stage breast cancer than non-Hispanic whites. Given that recent observance of Martin Luther King Day, I thought that it would be particularly appropriate to address a leading example of health disparities.
A new NIH-funded study of more than 175,000 U.S. women diagnosed with breast cancer from 2010-2016 has found that nearly half of the troubling disparity in breast cancer detection can be traced to lack of adequate health insurance. The findings suggest that improving insurance coverage may help to increase early detection and thereby reduce the disproportionate number of breast cancer deaths among minority women.
Naomi Ko, Boston University School of Medicine, has had a long interest in understanding the cancer disparities she witnesses first-hand in her work as a medical oncologist. For the study published in JAMA Oncology, she teamed up with epidemiologist Gregory Calip, University of Illinois Cancer Center, Chicago [1]. Their goal was to get beyond documenting disparities in breast cancer and take advantage of available data to begin to get at why such disparities exist and what to do about them.
Disparities in breast cancer outcomes surely stem from a complicated mix of factors, including socioeconomic factors, culture, diet, stress, environment, and biology. Ko and Calip focused their attention on insurance, thinking of it as a factor that society can collectively modify.
Many earlier studies had shown a link between insurance and cancer outcomes [2]. It also stood to reason that broad differences among racial and ethnic minorities in their access to adequate insurance might drive some of the observed cancer disparities. But, Ko and Calip asked, just how big a factor was it?
To find out, they looked to the NIH’s Surveillance Epidemiology, and End Results (SEER) Program, run by the National Cancer Institute. The SEER Program is an authoritative source of information on cancer incidence and survival in the United States.
The researchers focused their attention on 177,075 women of various races and ethnicities, ages 40 to 64. All had been diagnosed with invasive stage I to III breast cancer between 2010 and 2016.
The researchers found that a higher proportion of women receiving Medicaid or who were uninsured received a diagnosis of advanced stage III breast cancer compared with women with health insurance. Black, American Indian, Alaskan Native, and Hispanic women also had higher odds of receiving a late-stage diagnosis.
Overall, their sophisticated statistical analyses traced up to 47 percent of the racial/ethnic differences in the risk of locally advanced disease to differences in health insurance. Such late-stage diagnoses and the more extensive treatment regimens that go with them are clearly devastating for women with breast cancer and their families. But, the researchers note, they’re also costly for society, due to lost productivity and escalating treatment costs by stage of breast cancer.
These researchers surely aren’t alone in recognizing the benefit of early detection. Last week, an independent panel convened by NIH called for enhanced research to assess and explore how to reduce health disparities that lead to unequal access to health care and clinical services that help prevent disease.
References:
[1] Association of Insurance Status and Racial Disparities With the Detection of Early-Stage Breast Cancer. Ko NY, Hong S, Winn RA, Calip GS. JAMA Oncol. 2020 Jan 9.
[2] The relation between health insurance coverage and clinical outcomes among women with breast cancer. Ayanian JZ, Kohler BA, Abe T, Epstein AM. N Engl J Med. 1993 Jul 29;329(5):326-31.
[3] Cancer Stat Facts: Female Breast Cancer. National Cancer Institute Surveillance, Epidemiology, and End Results Program.
Links:
Cancer Disparities (National Cancer Institute/NIH)
Breast Cancer (National Cancer Institute/NIH)
Naomi Ko (Boston University)
Gregory Calip (University of Illinois Cancer Center, Chicago)
NIH Support: National Center for Advancing Translational Sciences; National Cancer Institute; National Institute on Minority Health and Health Disparities
Posted In: News
Tags: African American health, Alaskan Natives, American Indian, black, breast cancer, cancer, cancer diagnosis, health disparities, health insurance, Hispanic, insurance, oncology, race, racial disparities, SEER, women's health
Cancer Metastasis: Trying to Catch the Culprits Earlier
Posted on September 15th, 2015 by Dr. Francis Collins
Caption: Scaffold of a cancer cell-attracting implant as seen by scanning electron microscopy.
Credit: Laboratory of Lonnie Shea
For many people diagnosed with cancer localized to the breast, prostate, or another organ, the outlook after treatment is really quite good. Still, most require follow-up testing because there remains a risk of the cancer recurring, particularly in the first five years after a tumor is removed. Catching recurrence at an early, treatable stage can be difficult because even a small number of new or “leftover” tumor cells have the ability to enter the bloodstream or lymphatics and silently spread from the original tumor site and into the lung, brain, liver, and other vital organs—the dangerous process of metastasis. What if there was a way to sound the alarm much earlier—to detect tumor cells just as they are starting to spread?
Reporting in Nature Communications [1], an NIH-funded research team from the University of Michigan, Ann Arbor, and Northwestern University, Evanston, IL, has developed an experimental device that appears to fit the bill. When these tiny, biodegradable scaffolds were implanted in mice with a highly metastatic form of breast cancer, the devices attracted and captured migrating cancer cells, making rapid detection possible via a special imaging system. If the results are reproduced in additional tests in animals and humans, such devices might enable earlier identification—and thereby treatment—of one of the biggest challenges in oncology today: metastatic cancer.
Tags: bioengineering, breast cancer, cancer, cancer detection, cancer diagnosis, circulating tumor cells, CTC, implantable scaffold, ISOCT, metastasis, metastatic cancer, oncology, PLG scaffolds, precision medicine, secondary tumors