Developments in AI for Cancer Care: April 2025 Roundup

Author

Oncology AI Trends Staff
April 30, 2025

🚀 8 MIN READ

Welcome to the fourth roundup of 2025. This April, collaboration is the name of the game, across academic institutions, established technology companies, and startups.

🔬RESEARCH CORNER

🌟 University Hospitals Cleveland Medical Center (UH) teams up with Qure.ai

  • UH is collaborating with AI HeathTech company, Qure.ai, to deploy the latter’s FDA-cleared chest X-ray AI technology, called qXR-LN.

  • This is to support earlier identification of lung cancers by way of spotting easy-to-miss early stage lung nodules.

  • The medical center will investigate in a clinical trial the AI algorithm’s ability to enhance the detection of pulmonary nodules on chest X-rays, by essentially using it as a second read, following a radiologist’s read.

  • Read more here.

🌟 Cancer AI Alliance receives sizable funding to help further cancer cause

  • Google Cloud and AI2, the AI-research organization founded by the late Paul Allen, have each committed $10 million to the Cancer AI Alliance to help the fight against cancer with artificial intelligence.

  • According to PR Newswire, “Google Cloud will power planet-scale AI infrastructure and data analytics tools, while Ai2 will provide critical expertise in training large-scale models focusing on cancer research.”

  • The funds and resources will be used to develop, train, and scale AI models across cancer centers within the alliance that can analyze vast amounts of patient data while preserving privacy.

  • Read more here.

🌟 AI model to help detect colon polyps faster and more accurately

  • Researchers at Yeshiva University have developed a deep learning AI model to help detect polyps faster and enhance polyp segmentation in colonoscopy images.

  • The model, called PolypSEAG-Net, builds upon prior models called convolutional neural networks (CNNs) by improving polyp detection false positive rates, segmentation accuracy, and visual clarity on the most relevant polyp features.

  • Read more here.

🌟 AI-driven biomarker for detecting cancer cachexia

  • Researchers at Moffit Cancer Center have developed an AI-driven biomarker that could enable earlier detection of cancer cachexia, a condition typically marked by significant weight and muscle loss in some cancer patients.

  • The AI model works by analyzing CT scans to “quantify skeletal muscle mass using an algorithm trained on annotated images” and marrying that with patients’ lab results and electronic medical records to predict each patient’s likelihood of developing cancer cachexia.

  • Read more here.

🌟 AI model to help guide treatment selection in kidney cancer patients

  • Researchers at UT Southwest Center have developed an AI model that can accurately predict which patients with clear cell renal cell carcinoma (ccRCC) can benefit from anti-angiogenic therapies.

  • Only about 50% of ccRCC patients benefit from this class of therapy, so the model may help address an unmet need of identifying who might or might not benefit from therapy, as part of making treatment decisions.

  • Read more here.

🚀 PRODUCT CORNER

🔖 Breakthrough Device Designation for a pan-cancer AI-powered diagnostic

  • The FDA has granted breakthrough device designation to Paige PanCancer Detect, an AI-assisted diagnostics for detecting cancer across tissue and organ types. This pathway will help speed up the product approval process.

  • As pathology workload increases, Dr. David Klimstra, co-Founder of Paige, the AI company behind the technology, believes the technology could help “reduce time to diagnosis, leading to faster results for patients.”

  • Paige had previously received BTD designation for other portfolio products, including Paige Prostate Detect, for prostate cancer, and Paige Lymph Note, for breast cancer metastases.

  • Read more here.

🔖 Breakthrough Device Designation for a pancreatic cancer AI model

  • DAMO PANDA, an AI-powered model developed by Damo Academy, the research arm of Alibaba Group, has also been granted breakthrough device designation by the FDA.

  • The tool uses AI to identify cancerous lesions on CT imagery and achieved 92.9% sensitivity and 99.9% specificity for pancreatic ductal adenocarcinoma (PDAC) lesion detection.

  • Read more here.

🔖 AI algorithm integration to help speed up pathologist image analysis

  • DeepLIIF, a virtual staining AI algorithm developed at Memorial Sloan Kettering Cancer Center by Saad Nadeem, PhD and Travis J. Hollman, MD, PhD, will be integrated into PathPresenter.

  • This is an image management platform for oncology research, thereby enabling pathologists to analyze complex pathology images more accurately and efficiently.

  • Read more here.

AI coming to the rescue of the Prior Authorization process

  • RISA Labs has raised $3.5M to help hospitals and healthcare systems eliminate prior authorization (PA) delays in oncology care through automation.

  • RISA’s platform called Business Operating System as a Service (BOSS), uses intelligent agents including large language models (LLMs) and digital twins to complete complex workflows related to PA more efficiently.

  • Read more here.

Three-way deal to set the stage for future AI-assisted discovery efforts

  • Tempus AI has entered into a three-way deal with AstraZeneca and Pathos AI to develop a multimodal foundation model that its partners can use to further their R&D objectives.

  • The model is intended to “gather biological and clinical insights, discover novel drug targets, and develop therapeutics for the broader oncology community.”

  • Read more here.

That’s all for the April edition! Till next time ✨