CT scan (Computed Tomography) is basically an X-ray image, which can be used to provide clinicians with a detailed view of our internal organs, and can usually diagnose various forms of cancer. Previously, the use of CT to diagnose liver cancer was hindered to some extent by changes in the shape and structure of individual livers and the similarity of tissues in adjacent organs in CT images.
Amita Das of the Technical Education and Research Institute of the Department of Electronic and Communication Engineering at Siksha’O’Anusandhan University in Orissa, as well as Electrical Engineering at SCB Medical School and DY Patil Ramrao Adik Institute of Technology in Nerul, New Mumbai The department has developed a new texture analysis technology based on adaptive fuzzy clustering, which can be used to classify abdominal CT scans to diagnose liver cancer. This method is based on the extracted texture, morphology, and statistical features from the scan and using them as input, which can judge the neural network classifier to distinguish between benign and malignant liver tumors.
Today, they tested their method with a series of 45 images and studied sensitivity, specificity, and accuracy. The team was able to achieve nearly 99% accuracy in detecting tumors, and this result is already very good. The researchers’ next plan is to provide more data and training for the system, thereby further improving the reliability of the technology and developing an automatic diagnostic method that does not have the possibility of human error.