Chinese scientists have developed a breakthrough model using artificial intelligence (AI) technologies to aid in the rapid and effective detection of lymph nodes in cancer treatment.
Human colorectal cancer cells
The researchers developed the AI-assisted screening model using MRI image data from 293 patients with rectal cancer. The subjects of the study sought treatment at the Sixth Affiliated Hospital of Sun Yat-Sen University from 2013 to 2016.
The research transformed detection experiences into algorithms. The AI was engineered to detect lymph nodes at only 1.37 seconds for each sample. This rate is 100 times quicker than regular human detection.
Furthermore, the authors of the study tested the AI model in cancer patients at four medical establishments in Beijing, Guangzhou, Suzhou, and Guizhou. They then compared their results with four Chinese radiologists who specialized in gastrointestinal disorders.
In the conclusion of their results, they reportedly determined that the AI model could accurately identify three-millimetre lymph nodes with a detection accuracy of 80 per cent.
The full findings of the study were published in the journal EBioMedicine under The Lancet. Moreover, researchers from the University of Wisconsin School of Medicine and Public Health were also involved in the study.
Read Also: Understanding AI: Artificial Intelligence and How it Works Made Easy
What Happens When Cancer Spreads Into the Lymph Nodes?
Lymph nodes are the first line of defence in the human immune system. It protects people from illnesses and viral infections. In the human body, lymph nodes are hundreds of small, round glands that gather in the neck, armpit, abdomen and groin.
Cancer that starts in another part of the body and spreads to the lymph nodes or other organs is called metastasis. When this occurs, it is important for clinicians to judge the growth of cancer.
Healthline explains that if cancer cells are only detected in the lymph nodes close to the original tumor, it may mean that the cancer is still at an earlier stage and has not spread far beyond its area of origin.
On the other hand, if cancer cells are found to have migrated to lymph nodes far from the initial tumor, it could indicate that cancer may be spreading at a faster rate. Furthermore, it suggests that the disease could be in a later stage.
A Fast and Accurate Alternative
Current MRI screening methods for detecting cancer are time-consuming. Additionally, it cannot identify all the lymph nodes in the scan regions, which gives off lower detection accuracy.
According to Gao Xin, the lead researcher from the Suzhou Institute of Biomedical Engineering and Technology, a single lymph node takes up less than one-thousandth of an MRI sample. Moreover, it is easily confused with blood vessels and other tissues.
He says that even an experienced radiologist has to spend three to ten minutes analyzing lymph nodes on one sample. Furthermore, he added how each case needs approximately 20 samples.
For small lymph nodes with a diameter of less than 5 mm, Gao says that the detection accuracy is often less than 70 per cent.
The team believes that the AI-assisted screening model can save doctors a great deal of manual labor. More importantly, it can also improve clinical efficiency, which will benefit more patients, according to Gao.
Finally, the AI model is also capable of detecting metastatic cancers in other human organs or tissue, he added.
