Hyderabad researchers develop AI to detect eye cancer – Telangana Today

In the new study, published in the journal Cancers (October 2024), researchers said they improved ArMOR and tested its ability to detect retinoblastoma in a multiracial patient group or cohort.

Hyderabad: For the first time in India, researchers from Hyderabad have developed an artificial intelligence (AI)-based model to detect eye cancer, also known as retinoblastoma (RB).

Eye researchers at Hyderabad-based LV Prasad Eye Institute (LVPEI), Dr. Develop the first AI model called “Artificial Intelligence and Machine Learning in Eye Oncology, Retinoblastoma (ArMOR)” to diagnose eye cancer across all ethnicities with pinpoint accuracy.

In the new study, published in the journal Cancers (October 2024), researchers said they improved ArMOR and tested its ability to detect retinoblastoma in a multiracial patient group or cohort.

The ArMOR technology was developed with the help of researchers from the International Institute of Information Technology (IIIT-Hyderabad) and Bourntec Solutions Pvt Ltd. refined. The researchers used deep learning and geometric calculations to extract various features of retinal tumors such as tumor seeds, blood vessels, hemorrhages, etc., which an AI can identify.

“The AI ​​model demonstrated 97 percent accuracy in detecting retinoblastomas and 98 percent, 93 percent, greater than 99 percent, 94 percent and 93 percent in classifying tumors into the International Classification of Retinoblastoma groups A through E “, say the researchers in the study published in Cancers.

As AI/ML is increasingly being explored in the field of intraocular tumors, early detection of retinoblastoma (RB) is critical to optimizing treatment outcomes. “Therefore, we wanted to use AI/ML to develop a potential screening tool for RB and have previously demonstrated the feasibility of training an AI model to detect and classify RB from fundus images in an Asian-Indian cohort. Further in this work, we examined the model’s ability to detect and classify RB in a multiracial cohort. Despite the unequal frequency distribution between races, we identified opportunities for improvement and retrained the AI ​​model to detect and classify RB,” the researchers said.

“We started this journey by developing an AI model to diagnose retinoblastoma that has been validated in the Asian Indian population. However, we found that not all races can use the same AI model. Through this study, we have refined our existing AI model so that it can be used on all breeds regardless of fundus color,” says Dr. Swathi Kaliki, head of the OEU Institute for Eye Cancer, LVPEI and corresponding author of this article.