Liver cancer is a serious disease that can be difficult to detect early. When a patient’s symptoms raise a concern, physicians usually use medical images, such as CT scans or MRIs, to look for signs of cancer or abnormal growth in the liver. But sometimes, these images are challenging to read, and small tumors can be missed. That’s where artificial intelligence (AI) comes in.
A recent study looked at how AI is being used to help doctors find and understand liver cancer better.
What is AI?
AI stands for artificial intelligence, which means computer programs that can learn from data and make decisions or predictions. In medicine, AI can look at lots of information, such as diagnostic images of the body, lab test results and patient history, and find patterns that might show signs of disease.
How is AI used in liver cancer?
In liver cancer, AI can be used to:
- Find tumors: AI can scan medical images and spot areas that look unusual (this can indicate abnormal tissue growth or cancer cells), even if they’re very small.
- Measure tumors: It can outline the shape and size of a tumor, which helps doctors plan treatment.
- Predict risks: AI can estimate how likely it is that the cancer will come back or spread.
- Help choose treatments: Based on the available data, AI can suggest which treatments might work best for a patient.
What kinds of AI tools are there?
Researchers are using different types of AI tools to help study liver cancer. Each tool works in its own way, but they all have the same goal: to look at medical data and help doctors understand what’s going on inside the body.
Here are the main types:
Machine Learning (ML):
- Machine learning is a type of computer program that learns from examples. For liver cancer, it might look at thousands of patient records and medical images to learn what cancer looks like. Once it’s trained, ML can analyze new cases and predict whether someone might have cancer or how serious it is.
- Think of it like teaching a computer to recognize patterns, just like how you might learn to spot different dog breeds by looking at lots of pictures.
Deep Learning (DL)
- Deep learning is a more advanced kind of machine learning. It’s especially good at looking at pictures, such as CT scans or MRIs, and finding tiny details that might be hard for a human to see. These tools use a neural network, which is a system that works kind of like a human brain; it connects lots of pieces of information to make decisions.
- Deep learning is often used to find small tumors, measure their size and even tell if they’re likely to grow or spread.
Multi-Modal Systems
- “Multi-modal” means using more than one type of information at the same time. These systems combine medical images with lab test results, patient history and other health data. By looking at everything together, AI can give a more complete picture of the patient’s condition and medical needs.
- It’s like solving a puzzle; you need all the pieces to see the full image. These tools help doctors understand not just where the cancer is, but how it might behave and what treatments could work best.
What are the challenges?
Even though AI is helpful, it’s not perfect. It needs a lot of good-quality data to work well. If the data is missing or not diverse, AI can make mistakes. Also, many AI tools are still being tested and aren’t used in hospitals yet. Doctors still need to understand how AI makes its decisions, which isn’t always easy. While it’s still being developed, artificial intelligence is helping researchers learn more about liver cancer and how to treat it.
Reference:
Wang, L., Fatemi, M., & Alizad, A. (2024). Artificial intelligence techniques in liver cancer. Frontiers in Oncology, 14, 1415859. https://doi.org/10.3389/fonc.2024.1415859