New advances in cancer diagnosis and treatment leverage AI, IBM Watson, CRISPR, and even NASA tools to help detect and beat the disease.
Cancer researchers are leveraging new tech tools to develop better options for diagnosis and treatment. AI, IBM Watson, nanorobots, and gene editing capabilities all have the potential to revolutionize the way we fight the disease. Here are 10 ways technology is enabling innovative treatment approaches.
1. Microsoft Project Hanover: Maximizing research
This week, Microsoft announced Project Hanover, which aims to use machine learning and natural language processing to help doctors sort through the vast amounts of new medical research published each year. This way, they can better determine the most effective and individualized treatments for each patient.
The system can automatically search through every published research study to find the most relevant information for a specific patient. Eventually, the team will use a Microsoft Azure cloud computing tool to allow doctors to model the treatments that may work best. Microsoft is also working with the Knight Cancer Institute at the Oregon Health and Science University to develop a machine learning approach to personalize treatments for acute myeloid leukemia, a complex, often fatal type of cancer.
2. NASA: Using space technology to find cancer markers
A NASA machine learning algorithm that identifies similarities between galaxies will now analyze tissue samples for signs of cancer. Earlier this month, NASA's Jet Propulsion Laboratory and the National Cancer Institute renewed a research partnership through 2021 to collect research on these biomarkers into one searchable network. This way, physicians can compare, for example, a CT scan with an archive of similar images to search for early signs of cancer, based on a patient's demographics. Ultimately, this could translate into new techniques for early diagnosis of cancer or cancer risk.
Dozens of institutions, including Dartmouth College's Geisel School of Medicine, Harvard Medical School's Massachusetts General Hospital, and Stanford's NIST Genome-Scale Measurements Group have joined the network. It is similar to NASA's Planetary Data System, in which all can share information.
3. Google DeepMind: Improving radiotherapy scans
In August, Google-owned DeepMind Health partnered with University College London Hospital (UCLH) to improve radiotherapy scans used to detect head and neck cancers. These CT and MRI scans are complicated, and usually take physicians about four hours to complete, in order to create a detailed map of the areas of the body that need to be treated with radiotherapy, and the areas of healthy tissue they must avoid.
But DeepMind can use machine learning to assist clinicians in this process with an algorithm that automatically identifies cancerous and healthy cells. It may be able to cut the time needed for an accurate read to just one hour, leaving doctors more time to spend on patient care, education, and research.
4. IBM Watson: Analyzing research and developing treatment
IBM Watson is also using machine learning with its Watson Oncology application to help oncologists analyze research and develop treatment plans. Clinicians and analysts from Memorial Sloan Kettering Cancer Center in New York have been working with the company since 2014 to train Watson to "interpret cancer patients' clinical information and identify individualized, evidence-based treatment options," according to the hospital. This involves interpreting physician notes, lab results, and the latest clinical research.
Oncologists in any location could access Watson and make faster, more specific treatment decisions, the hospital stated. Watson can also allow doctors to better match patients to clinical trials. The system, which can be accessed on a tablet, is already in use in hospitals in India and Thailand.
5. CRISPR: Inactivating cancer mutations
In a study published earlier this month in the Journal of the National Cancer Institute, researchers used CRISPR technology to diagnose and inactivate cancer mutations. CRISPR, or clustered regularly interspaced short palindromic repeats, refers to a naturally occurring and very precise genome editing tool made of DNA.
More than 500,000 cancer mutations currently exist. Researchers from the National Center for Tumor Disease Dresden, the German Consortium for Translational Cancer Research, and the Medical Faculty of the TU Dresden found that more than 80% of these mutations could be targeted and cut out with the CRISPR system without harming healthy cells.
The approach could also improve cancer diagnostics by specifically identifying mutations that lead to cancer growth, and then developing an individualized treatment, according to a press release.
6. AI: Interpreting mammogram results
Researchers from Houston Methodist Cancer Center developed AI software that interprets mammogram results 30 times faster than a human can, with 99% accuracy. In a study published last month in the journal Cancer, researchers demonstrate how the software quickly and intuitively translates patient charts into diagnostic information.
About 12.1 million mammograms are performed annually in the US, with 50% yielding false positives, according to the American Cancer Society. About 20% of the 1.6 million breast biopsies performed each year are not necessary. This technology can cut down on the time it takes doctors to interpret results, and help them more accurately assess patient cancer risk and the need for more testing.
7. Nanorobotics: Attacking cancer cells
Researchers from Polytechnique Montréal, Universite de Montreal, and McGill University created nanorobotic agents that can travel through a patient's bloodstream to attack cancer cells in tumors with medication. The study, published in Nature Nanotechnology, was performed on mice, which successfully received the nanorobotic agents in colorectal tumors.
"Chemotherapy, which is so toxic for the entire human body, could make use of these natural nanorobots to move drugs directly to the targeted area, eliminating the harmful side effects while also boosting its therapeutic effectiveness," said Sylvain Martel, director of the Polytechnique Montréal Nanorobotics Laboratory, in a press release.
8. MR imaging: Adjusting radiation in real time
ViewRay, Inc. recently received approval from China and Japan to market its MRI-guided radiation therapy machine, called MRIdian System. MRIdian uses MR imaging, cobalt radiation delivery, and intelligent software automation to provide high-quality pretreatment images. More importantly, it allows physicians to see soft tissue and adjust radiation doses in real-time, while the treatment is being delivered. This way, the doctor can align the tumor to the treatment beams and avoid other sensitive internal organs.
The machine does not expose the patient to additional ionizing radiation that is common with other such systems, according to a press release. It is the only system on the market that can both image and treat patients simultaneously, ViewRay stated.
9. Gene editing: Engineering immune cells
Scientists from the University College London Cancer Institute are using technology to engineer immune cells to improve their ability to kill cancer. In a study published in the journal Cancer Research in April, researchers used gene editing technology to alter the DNA inside the immune cells of mice, making them resistant to a tumor cell's ability to switch them off.
While cancer patients often use drugs called checkpoint inhibitors to block cancer cells from switching off immune cells that attack the cancer. However, these drugs impact all of a patient's immune cells, and often have undesirable side effects. If this research is successful in human trials, it could be another way for doctors to use a patient's own immune system to fight disease.
10. CIVO: Testing drug impacts
In 2015, researchers at the Fred Hutchinson Cancer Research Center and Presage Biosciences in Seattle created a device that can inject multiple drugs into tumors to test the effect of each drug and determine which will work best for treatment.
The device, called CIVO, involves up to eight needles loaded with drugs that are pressed into a tumor located close to the patient's skin. The needles are then removed, leaving behind a trail of each drug. A few days later, doctors remove a piece of the tumor, and examine the cells to see whether each drug killed tumor cells, slowed their growth, or had no effect. CIVO has been tested on mice, dogs, and some human lymphoma patients, who did not report any adverse effects.
- 3 ways to maximize your big data team's cognitive computing investment (TechRepublic)
- Microsoft's next big AI project? Helping 'solve' cancer (ZDNet)
- How 4 universities are using 3D printing to create ears, cartilage and blood cells (TechRepublic)
- IBM contributes Watson technology to Joe Biden's cancer moonshot initiative (ZDNet)
- Why Microsoft stored 200MB of data in DNA strands, and what's next (TechRepublic)