Dr. Abdallah highlights a predictive AI model for patients with kidney cancer

In this video, Nour Abdallah, MD, discusses an AI model developed to predict postoperative estimated glomerular filtration rate (eGFR) in patients with kidney cancer. The findings were presented at the 2023 ASCO Genitourinary Cancers Symposium in San Francisco, California. Abdallah is a postdoctoral research fellow at Cleveland Clinic Glickman Urological and Kidney Institute in Cleveland, Ohio.

Video Transcript:

Could you describe the background of this study?

Dr. Christopher Weight, MD, who is one of my mentors at the Cleveland Clinic, has a high interest in studying the use of artificial intelligence and how it can enhance the diagnosis and management of kidney cancer. We know that nephrectomy is one of the main treatments of localized kidney cancer, and the American Urological Association, the AUA, recommends having a partial nephrectomy over a radical nephrectomy whenever the postoperative predictive eGFR falls below the value of 45.

Dr. Weight thought that we could try, through an AI model, to predict in a fully automated way this post-operative eGFR. We wanted to know if this AI model can behave as accurately as previously validated clinical models and models based on semi-automated volumes of the kidney.

What were the main findings?

This model that we developed was able to behave as accurately as clinical models that were previously validated, whether to predict the post-operative eGFR in value or to predict the cutoff value which is 45. We feel that this AI model would benefit the patients in a way that does not need any clinician time. It doesn't need any clinical details. It just needs the preoperative eGFR value and the CT scan. We feel that it can be easily implemented in the healthcare system at the point of care to help the physician guide the patient towards the best surgery type for them.