Lung cancer is one of the most deadly and risky diseases in the world. The prediction of the type of lung cancer or its stages can now be possible by using computer models having a specific algorithm present in them. As the population being affected by the disease is very large, there is an urgent need for the diagnostic method.
Hence, the researchers at the UT Southwestern Medical Center have come up with the computer algorithm which can help the doctors treat the patients on the basis of their risk at the proper stage and time. The algorithms were feed into the computer after a detailed study of various lung cancer tissue samples taken by the pathologists in order to help them differentiate between the malignant, benign or the type of cancer. Even though the data obtained for the algorithm is accurate proving the approach to be foolproof, but still it has to undergo a number of tests before being clinically used.
Speaking of the computer model, the researchers studied the non-small cell lung cancer and small cell lung cancer tissues. They analyzed about 3206 cancer tissue sample slides of 523 adenocarcinoma patients and 511 squamous cell carcinoma patients. These samples were provided by the “The Cancer Genome Atlas” which has details about the genetic, image, and clinical properties of cancer. The routine pathological method for the diagnosis and prognosis of cancer is a laborious and time-consuming method. Thus, the development of the computer model can help the pathologists perform their tissue specimen observation and prediction task at a very faster pace. The computer model has a pathological image scanner and a desktop that helps you see through the tissue samples. The device has been made very feasible and simple so that it can be used for clinical purposes.
The researchers are going to patent the process as soon as possible. The model helps study not only the structure and form of cells but also its microenvironment, which tends to play an important role in cancer. The scientists after studying the cell structure, size, color, texture, nuclei location, and shape have found out about 12–18 basic features which can help group the cells into two low-risk and high-risk groups.
One of the basic tests is to translate the images features identified by the computer into a clinical language which the pathologists can understand.
Hence, the computer image analysis technique when integrated with clinical tests or molecular analysis can prove to be the best option for the doctors during cancer tests.