For years, pathologists have been identifying cancer by looking at slides containing tissues stained with fluorescent dyes to make the malignant cells more visible. But, with the evolution of AI, this age-old process of identifying cancer tissues has become quicker and more accurate. AI has the potential to analyze and process heaps of data from various medical tests and predicts the prognosis of a patient and suggests doctors with various possible diagnosis and treatment options. AI’s application in cancer diagnosis is creating breakthroughs, but the technology is expected to undergo multiple changes before taking on the ultimate challenge—curing cancer. For example, a new AI-based intelligent computer has a convolutional neural network or CNN (an artificial network of nerves) that has the ability to identify skin cancer much more accurately than 58 dermatologists from 17 countries. Using machine learning, the device quickly evaluates the information presented to it and improves its ability to spot skin cancer cells.
In order to further improve the accuracy of cancer diagnosis, Weill Cornell Medicine, a biomedical research unit and medical school of Cornell University, and New York–Presbyterian, a nonprofit university hospital in New York City, have collaboratively developed an AI-based computer program that analyzes pathology images and determines whether or not a particular tissue is malignant and if the cell is cancerous, it then identifies the type of cancer. SOPHiA Genetics, a health tech company based in Switzerland, is also using AI to pinpoint gene mutations behind cancer and assist doctors in prescribing the best treatment. The AI takes raw genome data and studies it to decipher the molecular profile of a person’s cancer cell and find more suitable and personalized treatment options.
Although these AI initiatives are still at their early stage, companies in the oncology space that integrate AI as an essential part of their drug discovery and development process will gain a significant advantage over other competitors in the market. AI algorithms can be used to identify patterns hidden in large volumes of data thereby highlighting key differences between diseased and healthy cells. These companies will experience various advantages that AI presents—lower costs for the drug development timelines, more efficient identification of drug targets, and enhanced patient stratification methods. All of these factors will ultimately lead to larger revenue growth and more effective therapies being introduced quickly to the oncology market in the years to come.