A new AI (AI) application that can diagnose endocrine cancer is speed and accuracy and is presented on Sunday at Endo 2025, the annual general meeting of the Endocrine Society, held in San Francisco, California.
Research published by Jansi Rani Sethuraj, BSN, RN and CCRN of the University of Texas Health Sciences Center in Houston implements universally accessible and computationally efficient AI applications. The AI application is intended to democratize professional-level cancer diagnosis and makes it available on basic internet-connected devices, including smartphones.
Endocrine cancer affecting organs such as the thyroid, ovaries, pancreas, pituitary gland, and adrenal glands poses unique challenges due to their complex hormonal effects and difficult diagnostic profiles. An estimated 10 million cancer-related deaths each year bring about the need for innovative and scalable diagnostic solutions. This new AI-powered tool leverages efficient and advanced deep learning architectures such as ResNet to analyze a wide variety of medical data, including computerized tomography (CT) scans, magnetic resonance imaging (MRI), ultrasound (USG), and histopathology imaging, enabling comprehensive cancer detection.
According to Sethuraj, the AI model demonstrates exceptional diagnostic accuracy, reportedly exceeding 99% on specific validation datasets across multiple endocrine cancer types. These results are consistent with recent studies showing that AI can achieve high accuracy in classifying endocrine tumors, but actual performance may differ.
Two researchers, Ramya Elangovan and Kavin Elangovan of AIM Doctor in Houston, Texas, curated an anonymous endocrine cancer imaging dataset representing a diverse population across six continents. These images were used to train and validate deep learning models that could detect and stage multiple endocrine cancers with extremely high accuracy. The reliability and ease of use of the application has been independently assessed by health professionals from multiple international organizations, highlighting the potential for global applicability. The streamlined application design allows for rapid image analysis, allowing devices with limited computing resources to process each image in under a second.
By providing clinicians and primary care providers with access to expert-level diagnostic support everywhere, this technology has the potential to reduce diagnostic errors, accelerate treatment decisions, and improve patient outcomes worldwide, particularly in resource-limited settings.
“By democratizing access to advanced diagnosis, this AI innovation presents a paradigm shift in cancer care, providing early detection, more accurate treatment, and better survival for patients facing endocrine malignancies.” “This AI-powered application can provide fast, reliable, affordable endocrine cancer diagnosis anywhere, and for anyone, thereby closing the gaps in cancer treatment and increasing health equity around the world.”
