AI Revolutionizes Agricultural Health

Pancreatic cancer is a relatively rare disease that affects about 1 in 67 men and 1 in 60 women, according to the American Cancer Society. It is the 12th most common cancer worldwide and, unfortunately, is notorious for its late diagnosis and poor survival rates.

This raises great concern for farmers and those who work in the agricultural industry. Exposure to harmful herbicides and pesticides is a common denominator, along with exposure to diesel fumes, genetic factors and other risks.

To address this, researchers with The Ohio State University Comprehensive Cancer Center — Arthur G. James Cancer Hospital and Richard J. Solove Research Institute (OSUCCC - James) — are advancing early detection of precancerous cysts and lesions with the use of artificial intelligence (AI).

Dr. Somashekar Krishna, a gastroenterologist at OSUCC, and his team are pioneering this technology, working to revolutionize the way pancreatic cancer is detected and diagnosed in hospitals and clinics worldwide.

Pancreatic Cysts and Farmers

In the early stages, approximately one-fourth of all pancreatic cancers begin as cysts, making it challenging to determine their potential for developing into cancer.

What Are Pancreatic Cysts and Why Do They Matter?

The pancreas is an organ in the belly that helps digest food and controls blood sugar. Small fluid-filled sacs, called pancreatic cysts, sometimes form in the pancreas.

“These cysts are often found by accident when an individual has a CT (computed tomography) scan or MRI (magnetic resonance imaging) for another health issue,” Krishna explained.

“While many cysts are harmless, some can turn into pancreatic cancer over time.”

To determine whether a cyst is safe or dangerous, doctors use endoscopic ultrasound with needle-based confocal laser endomicroscopy (EUS-nCLE).

“This is a fancy way of saying that a tiny camera and laser microscope are inserted through a thin tube into the stomach to look closely at the cyst. This gives doctors a real-time, zoomed-in view of the cells inside the cyst, helping them determine if it is cancerous or not.”

Krishna continued, stating that early detection is critical because pancreatic cancer often has no symptoms in its early stages.

“By the time a person experiences weight loss, belly pain or jaundice (yellowing of the skin), the cancer may already be advanced,” he added.

For farmers, exposure to agricultural chemicals and other risk factors dramatically increases the likelihood of pancreatic cancer.

Farmers: A High-Risk Population

The Agricultural Health Study, funded by the National Cancer Institute and the National Institute of Environmental Health Sciences, has studied cancer and other health issues in farming populations since 1993.

More than 89,000 farmers and their spouses in Iowa and North Carolina have participated in the study, providing invaluable information on how agriculture, lifestyle and genetic factors impact their health.

In one study, findings showed that pancreatic cancer has been linked with pesticide and herbicide use, connecting a unique association with two specific chemicals — pendimethalin and S-Ethyl dipropylthiocarbamate (EPTC).

“Farmers may be at higher risk for pancreatic cancer due to long-term exposure to pesticides, herbicides and possibly from diesel fumes from tractors and machinery,” Krishna shared.

“These chemicals can cause inflammation and damage to the pancreas over time, increasing the risk of cancer.”

To treat their fields, farmers will often use weed-killing herbicides, such as pendimethalin and EPTC, to remove competing vegetation such as grasses and weeds from around crops.

A report, published in the International Journal of Cancer, found that exposure to these agricultural chemicals increased cancer risk significantly. The use of pendimethalin, for example, increased the risk by 40% among farmers with moderate exposure.

In contrast, the chemical EPTC was linked to an 80% increased risk with just low-level exposure. For those heavily exposed, the risk rose to two and a half times that of pesticide applicators who had never used this chemical.

On top of chemical exposure, Krishna said farmers face other risk factors, including:

• Breathing in diesel fumes from tractors and heavy machinery.
• Chronic inflammation of the pancreas from chemical exposure.
• A diet high in processed meats and fats, which may contribute to inflammation.

“Although we can’t always avoid these risks completely, wearing protective gear, using safer farming methods and getting regular medical check-ups can help reduce the danger.”

AI Technology Advances Detection & Diagnosis

Analyzing Virtual Biopsies

Typically, when doctors perform endoscopy-guided microscopy, they see live microscopic images of the pancreas. These images move quickly, making it hard to analyze in real time, even for trained specialists.

“While expert doctors can detect warning signs, not all doctors have the same experience, and interpretations can vary,” said Krishna. “This is where AI can help. AI can analyze the images automatically, finding patterns and early warning signs that might be missed.”

This technology, according to Krishna, makes diagnosing pancreatic cysts more accurate, faster and consistent across different hospitals and doctors. It can also help doctors decide if surgery is necessary or if the cyst can be safely monitored.

Accurate Detection & Diagnosis

Typically, when a pancreatic cyst is found, the doctor removes some fluid from the cyst using a needle to test it.

“This method, however, is only 60% to 70% accurate,” Krishna explained. “This means some dangerous cysts might be missed, while some harmless cysts may lead to unnecessary surgeries.”

EUS-nCLE has a much higher accuracy (over 95%) because it allows doctors to see the cyst in real-time at a microscopic level instead of just testing fluid.

Adding AI improves the process by:

• Making diagnosis faster and more accurate.
• Reducing differences in opinions between doctors.

“This could change the way pancreatic cysts are diagnosed worldwide, making early detection of pancreatic cancer easier and more reliable.”

According to Krishna, with AI, doctors might be able to detect cancerous changes months or even years earlier than before. This means that individuals at high risk, like farmers, could be monitored more closely and treated sooner if needed.

A Message to Farmers

Krishna urges farmers to stay proactive and know the risk factors associated with pancreatic cancer:

“If you have a family history of pancreatic cancer, exposure to pesticides or chronic pancreatitis, talk to your doctor about early screening.”

He also recommends monitoring and watching for symptoms such as unexplained weight loss, belly pain, jaundice or a new diagnosis of diabetes.

“Consider advanced screening if you have a pancreatic cyst. Not all cysts are dangerous, but advanced tools like EUS-nCLE and AI-assisted imaging can help determine if you need treatment. And stay proactive about your health. Regular check-ups and early detection can make all the difference,” added Krishna.