Liver cancer is one of the most serious forms of cancer in the world. The most common type is called hepatocellular carcinoma, or HCC.
It often develops quietly and is usually found at a late stage, when treatment is more difficult. Because of this, early detection is very important for improving survival.
A new study published in the journal Cancer Discovery has found that artificial intelligence may help doctors identify people at risk of liver cancer much earlier than before. The research was led by scientists from RWTH Aachen University and the Technical University of Dresden.
Currently, doctors mainly screen people who already have severe liver disease, such as cirrhosis. However, many people who develop liver cancer do not have a known history of these conditions. This means that a large number of at-risk individuals are not being monitored.
To address this problem, the researchers developed a machine learning model that can predict liver cancer risk using simple and commonly available medical data. This includes basic information such as age and sex, medical history from electronic health records, and routine blood test results.
The study used data from the UK Biobank, which contains health information from more than 500,000 people. Among these individuals, 538 cases of liver cancer were identified. Interestingly, about 69 percent of these cases occurred in people who had not been previously diagnosed with major liver disease.
The researchers trained their model using most of this data and then tested it on the remaining portion. They also validated the model using another large dataset from the United States, called the All of Us registry, which includes more diverse populations.
The model used a method called a random forest, which combines many small decision-making steps to arrive at a final prediction. This approach helps improve accuracy and reliability.
The results showed that the model performed very well. It achieved a high level of accuracy in identifying individuals at risk of liver cancer. Importantly, the best-performing version of the model only needed simple clinical data and did not require complex or expensive genetic testing.
In fact, adding advanced data such as genomics did not significantly improve the model’s performance. This means that the tool could be widely used in everyday healthcare settings, including in areas with limited resources.
The researchers also compared their model with existing clinical tools used to estimate liver cancer risk. These include scores based on liver function and fibrosis. The new model outperformed these traditional methods by identifying more true cases while reducing false alarms.
To make the model more practical, the researchers simplified it further. Even when using only about 15 common clinical features, it still performed better than existing tools.
This study has several strengths. It uses large datasets and includes validation in different populations. It also focuses on practical application, making it easier for doctors to use in real-world settings.
However, there are also limitations. The study is based on past data and does not prove how well the model will work in future clinical use. In addition, some important risk groups, such as people with viral hepatitis, were not strongly represented.
Overall, the findings are promising. They suggest that simple AI tools could help doctors identify people at risk of liver cancer earlier and guide them toward screening and treatment.
In conclusion, this research highlights the potential of artificial intelligence to improve early cancer detection. By using routine health data, it may be possible to find high-risk patients who would otherwise be missed, leading to better outcomes in the future.
If you care about liver health, please read studies about simple habit that could give you a healthy liver, and common diabetes drug that may reverse liver inflammation.
For more information about health, please see recent studies about simple blood test that could detect your risk of fatty liver disease, and results showing this green diet may strongly lower non-alcoholic fatty liver disease.
Source: RWTH Aachen University.
