The National Cancer Institute estimates that 1,735,350 new cases of cancer will be diagnosed in 2018. With this year more than halfway over, it is assumed that nearly one million new cases of cancer have been diagnosed by this point.
Cancer is a complex disease that develops differently for every individual diagnosed. Some cancers are more treatable than others, and there are a variety of factors that play into how one goes about being treated. However, one technology that may universally be able to help every patient is artificial intelligence (AI). With the ability to analyze data from a plethora of sources and capacity to learn from prior mistakes, AI has the potential to revolutionize diagnostics for a range of different diseases.
What is Artificial Intelligence?
Artificial intelligence is a term that refers to machines that are programmed to mimic human reasoning. The goal of AI is to learn from failure and be able to provide the best recommendation for a specific subject. Whether it be solving equations, finding the best treatment options for cancer patients, or aiding in vehicle autonomy, AI is becoming a powerful resource across multiple industries.
To better understand AI, we must take a look at some examples of how it is being applied to modern-day problems.
AI and Cancer Research
AI is beginning to be accepted into a variety of different fields, one of them being health care. When hearing that AI is being adopted into this industry, we often think that human-like robots are going to be performing surgery. While this could be something that happens in the future, it is not how AI is being used currently. Through supercomputers like IBM Watson, hospitals are beginning to utilize this technology. Watson has partnered with the Memorial Sloan Kettering Cancer Center and other clinics to develop an oncology computing system that will be used to advance cancer research.
Watson is currently analyzing 13 different cancer types, including breast cancer. According to IBM, “Mayo Clinic physicians presented a poster presentation at the ASCO Annual Meeting, reporting that Watson for Clinical Trial Matching boosted enrollment in breast cancer trials by 80% following implementation (to 6.3 patients/month, up from 3.5 patients/month in the immediate 18 months prior).” This means that Watson is qualifying more patients to enter clinical trials and potentially saving their lives, if the trial is effective.
Recently, Chinese researchers developed another great AI system to determine if collected prostate tissue samples were cancerous. This system has shown a successful amount of accuracy, with some claiming it is as accurate as a pathologist. The system can also detect and classify the malignancy of the cancer. Using this system, a study that utilized 918 prostate tissue samples from 283 different patients were tested for abnormalities. The results showed that 99.83 percent of diagnoses using this method were accurate. While this is amazing progress, Hongqian Guo, the urologist leading the study stated, “We still need an experienced pathologist to take responsibility for the final diagnosis. What it will do is help pathologists make better, faster diagnoses, as well as eliminating the day-to-day variation in judgement which can creep into human evaluations.”
Future Applications of AI
With AI being used for cancers as complex as breast and prostate, we will hopefully see AI aid in diagnoses of even rarer cancers. By detecting cancer earlier and more accurately, prognoses of other complicated cancers, such as mesothelioma and ovarian cancer, could greatly improve.
Like any other new technology, there have been criticisms as to how accurate it is and an expectation that it should work immediately. However, machine learning takes time and is constantly improving in order to provide better recommendations. As the technology becomes more advanced and researchers provide better data for it to learn from, the possibilities will be endless. Hopefully, we will see AI utilized in more industries, and help improve more lives.