Back

Automated imaging-based prognostication (IPRO) for stage I non-small cell lung cancer using deep learning applied to computed tomography (CT) scans.

|
June 2, 2022

Toronto, ON | Chicago, IL – Altis Labs, Inc. (“Altis”), the computational imaging company accelerating clinical trials with AI, is announcing the results of applying IPRO to stage I NSCLC patients will be presented at this year’s American Society of Clinical Oncology conference. 

IPRO, the fully-automated deep learning system developed by Altis, processed pre-treatment CT scans of stage IA and IB NSCLC patients to generate prognostic predictions beyond tumor, node, metastasis (TNM) sub-stages.

According to IPRO, the 20% of patients predicted to have the highest mortality risk had significantly lower median survival (48.0 months vs. 69.5 months) compared to the 20% of patients predicted to have the lowest risk.

This work not only highlights IPRO’s prognostic accuracy compared to subjective, reductionist staging guidelines like TNM, but also shows promise for enabling better patient stratification and quantification of treatment effect in populations that require long-term follow-up to measure therapeutic efficacy.

Felipe S. Torres, MD, PhD, cardiothoracic radiologist from the Joint Department of Medical Imaging at Toronto General Hospital, was the lead author of the study.

Read the full abstract here.

About Altis Labs

Altis Labs is the computational imaging company advancing precision medicine. We believe that medical imaging is the richest source of untapped clinical insight. Biopharma companies use our software platform, Nota, to accelerate and de-risk all stages of clinical development. Trained on over 182 million real-world images with associated diagnostics, treatment information, and outcomes, AI models hosted on Nota predict clinically meaningful outcomes from baseline and follow-up scans to more accurately stratify patients and quantify treatment effect. Altis is proudly based in Toronto, Canada.

To learn more, email info@altislabs.com.