According to The Print, Johns Hopkins Kimmel Cancer Center researchers utilized AI to diagnose lung cancer successfully. They correctly detected more than 80% of liver cancers in a recent experiment with 724 patients.
The study’s findings were presented at the American Association for Cancer Research’s Special Conference on Precision Prevention, Early Detection, and Interception of Cancer and published in Cancer Discovery.
Cell-free DNA is DNA released by cancer cells and reaches the bloodstream. The DELFI (DNA assessment of fragments for early interception) blood test looks for changes in DNA fragmentation (cfDNA). DELFI technology was used to analyze blood plasma samples from 724 persons in the United States, Europe, and Hong Kong. The results revealed that a significant proportion of the participants had hepatocellular carcinoma (HCC).
The researchers think theirs is the first attempt to independently validate a genome-wide fragmentation analysis in two high-risk groups, which include persons of diverse racial/ethnic origins whose liver tumors are caused by different causes.
Cirrhosis caused by chronic liver diseases such as chronic viral hepatitis or non-alcoholic fatty liver disease puts about 400 million people at risk of developing HCC.
According to Victor Velculescu, MD, Ph.D., professor of oncology and co-director of the Cancer Genetics and Epigenetics Program at the Johns Hopkins Kimmel Cancer Center, more excellent early detection can reduce liver cancer fatalities. Current cancer screening technologies, however, are overused and frequently miss significant cancers.
According to Foda, 75 of the total 724 plasma samples acquired in the US and EU were from people with HCC, and the machine learning model was built and evaluated using this data. Machine learning, a type of artificial intelligence, mixes data and algorithms to enhance precision. To corroborate the findings, 90 people with HCC, 66 people with hepatitis B virus (HBV), 35 people with HBV-related liver cirrhosis, and 32 people with no known risk factors had their plasma samples examined.
The DELFI approach may disclose how DNA is packed inside a cell’s nucleus from simple blood test results by examining the amount and location of cell-free DNA present in circulation from different sections of the genome.
Normal cells have DNA packed so that the various portions of the genome remain separate. Cancer cell nuclei, on the other hand, appear to have opened a suitcase, the contents of which have been spread over the genome. Cancer cells release random pieces of DNA into the body’s circulation once they die.
DELFI can identify the presence of cancer by evaluating millions of cfDNA fragments for abnormal patterns such as size and amount of DNA at distinct genomic locations. The authors of this study argue that the DELFI approach is better than other sequencing methods for screening since it requires less sequencing coverage.
It has been established that cfDNA fragments isolated from plasma samples may be utilized to classify lung cancer correctly. DELFI was developed by examining how different types of fragmentation manifested in each sample.
The DELFI approach effectively identified early-stage liver tumors in people with a moderate risk of acquiring the illness, with an overall sensitivity of 88% and a specificity of 98%. The test was 85% sensitive and 80% specific in samples from high-risk people.
