Google Button Μake us preferred on Google

Pancreatic cancer remains one of the biggest challenges in modern oncology. In recent decades there has been rapid progress in surgical techniques, chemotherapy protocols, and the overall care of patients before, during and after surgery. Even so, the prognosis is still poor and five-year survival rates remain low.

A ‘Silent Threat’

The disease often goes undiagnosed in its early stages, when it can still be controlled. This is mainly because of the way these tumors develop. They cause few or no symptoms, which makes them a “silent threat.”

Unfortunately, international research has shown that late diagnosis is linked to several problems:

  • Surgery that is technically more demanding.
  • More complications.
  • Higher rates of metastasis or recurrence.
  • Lower chances of a cure.

Against this complex background, artificial intelligence is emerging as a set of powerful new tools. They promise to substantially improve how we approach the diagnosis and treatment of pancreatic cancer.

Artificial intelligence refers to computer systems that can analyze large and complex datasets, recognize patterns, and support decision-making as a useful assistant to doctors. In medicine, this data comes from many sources:

  • Imaging scans.
  • Tissue samples and pathology reports on surgical specimens.
  • Lab results.
  • Electronic and paper patient records.

AI can combine and process this information with a speed and accuracy that often exceed human analysis, but it does not replace it. Instead, it works alongside clinical judgment and strengthens it. It provides useful information that plays a decisive role in key decisions about planning personalized treatment.

Put simply, the algorithm combines images and data from hundreds of thousands of patients and “sees” lesions before the human mind can detect them. An AI system could be compared to a team of highly skilled, experienced, tireless doctors with perfect memory.

These systems learn very quickly, have an almost limitless memory and don’t suffer from human fatigue. That makes them a valuable assistant. Once doctors have weighed their suggestions clinically, those suggestions help improve patient care.

The ‘Second Reader’ And Early Diagnosis

What about early diagnosis, which is the key to a better prognosis? In recent years, computer models have been developed to analyze imaging scans such as CT and MRI. They can accurately spot small changes in the pancreas that may escape even experienced doctors. In many cases, AI acts as a “second reader,” making diagnosis more sensitive and reducing the risk of error.

Of particular interest are studies showing that such systems can detect early signs of the disease in scans taken months before the clinical diagnosis.

The algorithm also makes an important contribution to endoscopic ultrasound, an invasive procedure that is key to confirming a cancerous lesion through tissue sampling. By analyzing the images, AI systems can tell benign from malignant lesions with high accuracy, which makes the result less dependent on the examiner’s experience.

They can also guide biopsies in real time. This improves the quality of the samples and increases the chance of a correct diagnosis on the first attempt. It also helps avoid serious complications, such as bleeding and injury to surrounding tissue.

AI’s contribution is not limited to diagnosis. It also extends to treatment planning. Deciding whether a tumor can be operated on is a critical point in managing the disease. AI can analyze imaging scans in detail and assess how the tumor relates to nearby structures, giving a more objective and reproducible assessment. This helps doctors choose the most suitable strategy, whether that is immediate surgery or treatment before surgery.

A Valuable Assistant In The Operating Room

AI is starting to play an increasingly active role even in the operating room. Combined with laparoscopic or robotic surgery and augmented reality technology, it can give the surgeon a clearer and more detailed view of the anatomy. It does this by displaying critical structures on the screen and helping avoid complications.

These applications have started being used successfully in less complex operations. As the technology improves, they are expected to make more complex procedures significantly safer and more precise in the future.

Overall, AI appears to be moving from research into everyday clinical practice. Its most mature applications today are diagnosis through imaging and support for decision-making, and new applications are constantly being developed.

The main challenges concern the quality, diversity and reliability of the data used to train the computer models. Other issues include the transparency of algorithms and the need to train doctors in using the new technologies.

AI is becoming part of our daily lives and offering solutions to simple everyday needs. Bringing it into the clinical management of demanding diseases such as pancreatic cancer could substantially change their course.

Its role is to support doctors, under critical judgment, not to replace them. It equips doctors with tools that improve accuracy, reduce errors and allow more personalized care. The ultimate aim is to improve our patients’ survival and, above all, their quality of life.

Afroditi Fotiadou is a general surgery resident at Heidelberg University Hospital and a PhD candidate at the Medical School of the National and Kapodistrian University of Athens.

Nikolaos Arkadopoulos is a professor of surgery, director of the 4th Surgical Clinic at Attikon University General Hospital, and dean of the Medical School of the National and Kapodistrian University of Athens.

Ioannis Hatzaras is an associate professor of surgery at the Medical School of the National and Kapodistrian University of Athens.