Artificial intelligence (AI) has been behind much of the progress in the health care sector. It‘s particularly promising for helping physicians diagnose illnesses faster than they otherwise might. More specifically, AI algorithms can look for minute details in gigantic sets of data, helping them find early signs of an ailment and get patients started on the proper treatments.
The World Health Organization (WHO) calls cancer a leading cause of death worldwide. However, it notes that early detection and screening are among the best ways of preventing the worst outcomes. Many treatments are more successful when used at an early stage of cancer, for example.
Here are some fascinating ways AI can help in improving cancer screening methods and detection rates.
1. Offering Non-Invasive Screening
Some current cancer detection methods are uncomfortable and awkward for patients. Despite many people tolerating that reality and getting screened at the recommended times, others delay those tests due to what they involve. AI could reduce that trend by alerting physicians to a patient’s cancer risk before putting them through the usual screening procedures.
Research done at University College London involved teaching AI to screen for esophageal cancer by determining the most common risk factors. More specifically, the algorithm examined data from 1,299 patients with a condition called Barrett’s esophagus, which causes abnormal cell growth in that part of the body. It’s not cancer but is the only known precursive factor to a person eventually developing it.
The AI concluded that eight factors made a person more likely to develop Barrett’s esophagus or esophageal cancer. Some, like age and gender, are outside a patient’s control. However, since smoking and excessive waist size were also among the factors, these are points that could come up during discussions about reducing someone’s cancer risk.
Laurence Lovat, the lead author of a study on this approach, explained, We propose developing a simple tick-box questionnaire, identifying risks such as a large waist circumference, severe stomach pain, and duration of heartburn, which could be filled out by a GP or by a patient using a mobile phone app.
He continued, The results would identify a high-risk group of people at an early stage who could then go on to have clinical screening to diagnose and treat esophageal cancer at a much earlier stage and significantly improve survival rates.
2. Providing a Radiation-Free Lung Cancer Screening Method and Better Treatment Methods
Statistics from the World Cancer Research Fund show that lung cancer is the most common cancer affecting men, and it ranks as the third most prevalent in women. Researchers developed an AI-based method that screens for lung cancer in the bloodstream by looking for DNA-shed fragments from cancer cell deaths.
They believe this method could be a game-changer in lung cancer detection because it does not expose people to radiation from CT scans, which is the usual way of checking patients for this health issue. Since only a small percentage of at-risk people receive CT scans to look for lung cancer, this new method could encourage more to come forward and do so sooner.
The AI studied samples from 800 people and was 94% successful in detecting lung cancer in various stages. There’s now a clinical trial underway with 1,700 participants, including healthy individuals and those with lung and other cancers.
This is an example of a so-called liquid biopsy method that gives people a non-invasive way to stay on top of their health. Another example from elsewhere highlighted how AI could help doctors provide accurate prognoses to people with cancer. That research concerned people with either lung cancer or dementia.
The tool used big data and AI to create a user-friendly precision-medicine platform for doctors to rely on during patient care. It compiles data and offers an expected prognosis, plus suggests possible interventions. In any case, getting information in one place from numerous sources could help doctors see the big pictures associated with their patients, reducing the chances that certain factors get missed.
3. Helping Less-Experienced Doctors Diagnose Skin Cancer
Besides taking preventive measures such as applying sunblock before going outside, wearing wide-brimmed hats outdoors, and avoiding the hottest part of the day, doctors also recommend that people check their skin for any changes and make appointments for professional attention if they find anything unusual.
AI-driven smartphone apps can also help people become aware of potentially cancerous moles, enabling them to connect to the right doctors sooner. Even the ones without AI are beneficial because they allow users to take pictures of their skin to refer to later rather than solely relying on memory.
As these AI apps for detecting skin cancer have become more prevalent, some people wonder if they might do even better than doctors at determining whether skin changes might have occurred because of cancer.
However, a 2020 study indicated that such AI applications perform best when they involve human input. More specifically, a team at the University of Queensland found that AI could successfully guide clinical decision-making for skin cancer diagnoses. However, the results were most accurate when people weighed in, too. Plus, AI provided the best help to clinicians who had comparatively fewer years of practice behind them.
Professor Monika Janda explained, Inexperienced evaluators gained the highest benefit from AI decision support, and expert evaluators confident in skin cancer diagnosis achieved modest or no benefit. These findings indicated a combined AI-human approach to skin cancer diagnosis may be the most relevant for clinicians in the future.
As people become more familiar with AI applications in health care, they often discover that the technology has both advantages and limitations. This is one such example that illustrates why artificial intelligence needs ongoing research.
AI Assists in the Cancer Fight
A cancer diagnosis is a life-altering event affecting patients and their loved ones. It also understandably fills most people with dread. However, due to AI applications like those listed here, people could soon feel more hopeful about the future of cancer detection and the most common screening methods.
Even if some of the options here and others like them don’t work quite as well as expected in the real world, efforts to see what AI can do well help doctors understand the best use cases for it and when traditional measures might offer better results.
