Iranian Company Develops AI-Based System for Accurate, Rapid Cancer Diagnosis
“We have launched a domain called Avax AI, which operates similarly to ChatGPT, but specializes in the field of medical image analysis. This system can detect the extent of cancer, determine the nature of tumors (benign or malignant) with an accuracy of more than 93 percent, and also analyze cardiovascular images,” said Mostafa Taqipour, the managing director of the knowledge-based company.
“The Avax AI project has been developed over 6 years and with an investment of nearly $4 million,” he added.
“Our software is continuously updated to increase the accuracy of the analysis and benefit from the latest scientific advances in the field of artificial intelligence,” Taqipour said.
In a relevant development in October, Iranian researchers in a joint research with their American counterparts had also succeeded in designing an artificial intelligence algorithm to detect cancerous tissues.
Sadeq Kazzemi Fard, the manager of this project, described the high accuracy of diagnosis as one of the advantages of this method, and said, “The artificial intelligence algorithm developed by identifying cancer tissue in the early stages enables doctors to prevent the spread of the disease.”
“Sina artificial intelligence model has been approved by the World Health Organization (WHO) and will be available to medical and research centers from the beginning of 2025,” he added.
“In its latest clinical trial, Sina artificial intelligence algorithm once again demonstrated its ability to accurately diagnose cancer and was able to identify a cancerous mass in the breast tissue of a young woman, which was mistakenly diagnosed as a cyst in the initial tests,” Kazzemi Fard said.
“This patient, who was diagnosed with suspected cancer in routine examinations, was accurately identified with the help of artificial intelligence and mammography of the cancer tissue in two places, and the disease was confirmed with a biopsy, and the treatment process began for the patient,” he explained.
4155/v
