Cancer can be a tough disease to spot. And this can make it a difficult disease to fight since your best chance for survival is to catch cancer early. Some cancer screenings are recommended for certain age groups and others are recommended for people at higher risk. But oftentimes it’s up to you to keep a close eye on your health and notify the doctor if anything unusual is going on.
If you do see something abnormal happening, doctors can use several diagnostic tools to figure out if it’s cancer. They’ll often start with a physical exam, feeling areas of your body to look for lumps that may indicate a tumor, or checking any change in skin color or enlargement of an organ that could indicate the presence of cancer. Urine and blood tests can also help doctors identify certain types of cancer, such as leukemia.
If the doctor does find something strange, the next step is often to get an image of the area through an ultrasound, X-ray, CT scan, bone scan, MRI or PET scan. A sample of any abnormal growths is then taken through biopsy, and the cells are delivered to a laboratory where doctors look at the sample under a microscope to identify cancer cells.
Usually, cancer cells look less orderly than normal cells, with varying sizes and lack of organization. But sometimes the cancer cells and healthy cells look similar under a microscope, which can lead to a misdiagnosis. Another problem with biopsy is that it’s time-consuming and expensive.
That’s why researchers in Singapore are exploring the use of a cutting-edge technology to identify cancer cells more quickly and cost-effectively…
Artificial intelligence, acidity and cancer cells
Another way of separating cancer cells from normal cells is to example the pH, or level of acidity, inside the cells. A team of researchers from the National University of Singapore (NUS) has developed a technique employing artificial intelligence (AI) to figure out whether a single cell is healthy or cancerous by analyzing its pH.
Compared with the several hours current methods of cancer cell imaging can take, each cancer test using this AI method can be completed in less than 35 minutes. And the AI method can classify single cells with an accuracy rate of more than 95 percent.
“The ability to analyze single cells is one of the holy grails of health innovation for precision medicine or personalized therapy,” says research lead Professor Lim Chwee Teck, Director of the Institute for Health Innovation & Technology (iHealthtech) at NUS. “Our proof-of-concept study demonstrates the potential of our technique to be used as a fast, inexpensive and accurate tool for cancer diagnosis.”
Current techniques for examining a single cell can cause toxic effects or even kill the cell. But the approach developed by the NUS research team can tell the difference between cells coming from normal and cancerous tissue, as well as among different types of cancer — all while keeping the cells alive.
In the NUS method, a pH-sensitive dye is applied to living cells that changes color depending on how acidic the cell is. Because of its intracellular activity, each cell displays a unique combination of red, green and blue (RGB) components under light. Cancer cells have an altered pH compared with normal cells, so they react differently to the dye, changing their RGB pattern.
After capturing the RGB components emitted from the dye inside the cells, the researchers used an AI-based algorithm to quantitatively map the unique acidic patterns so that the cell types examined can be identified easily and accurately. Thousands of cells originating from various cancerous tissues can be imaged at the same time, and single-cell features can be extracted and analyzed.
The NUS approach uses simple, inexpensive equipment and doesn’t require lengthy preparation to screen cells quickly and accurately. And because the method doesn’t cause toxicity or kill the cells, it would allow for further analysis down the line that could require live cells.
The researchers are planning to develop a real-time version of their technique where cancer cells can be automatically recognized and promptly separated for further molecular analysis, such as genetic sequencing, to find out if there is any potential mutation that can be treated with drugs. The team also plans to advance their concept to detect different stages of malignancies from the cells tested.
Warning signs of cancer
In many cases, cancer causes distinct symptoms that can help you catch it early on. Some things to watch out for include:
- Fatigue
- Night sweats
- Loss of appetite
- Recurring nausea or vomiting
- Unexplained weight loss
- New, persistent pain
- Blood in urine or stool, either visible or detected in special tests
- A recent change in bowel habits (constipation or diarrhea)
- Recurring fever
- Difficulty swallowing
- Chronic cough
- Changes in the size or color of a mole
- Changes in a skin ulcer or sore that does not heal
- A growth or mark on the skin that gets larger or changes in appearance
- Enlarged lymph nodes
- Lump in the breast
Some of these symptoms may be caused by less serious conditions, but it is always good to visit your doctor if you notice something, and to be sure and to keep your regular check-up appointments that may include screenings. Better to err on the side of caution than to let the symptom worsen and any possible cancer progress to the point where treatment may be difficult.
Sources:
NUS researchers harness AI to identify cancer cells by their acidity — National University of Singapore
Cancer — Mayo Clinic
Warning Signs of Cancer — Merck Manual