Developments in AI for Cancer Care: February 2025 Roundup

📰 The UK’s National Health Service (NHS) is launching the world’s largest trial that uses artificial intelligence to detect breast cancer. Over the next few years, AI will be put to the test across mammograms completed in England, to see whether it is as accurate at reading scans as a radiologist. Over 700,000 women will participate in the study. A positive study outcome may prove that with AI, the second reader system - whereby a radiologist checks mammograms for signs of breast cancer following an initial read by a different radiologist - may not be necessary to establish accuracy of diagnoses. This could free up radiologists to see more patients and help improve population health outcomes. Read more here.

📰 Researchers at Johns Hopkins University, led by Zongwei Zhou and Wenxuan Li, have rolled out the latest version of AbdomenAtlas, an “annotated public dataset containing abdominal CT scans from over tens of thousands of patients around the world.” AbdomenAtlas can be used to train early cancer detection AI models as well as ultimately help radiologists increase speed of diagnosis. Since its initial rollout in 2024, the team has continued to advance the dataset’s capabilities. Version 3.0 (officially AbdomenAtlas 3.0) is a dataset of over 9000 3D CT scans and includes organ-specific annotations such as pancreatic cancer staging. Read more here.

🚀 Manas AI, an AI-driven platform focused on revolutionizing drug discovery and development for cancer and other diseases, launched this month with a $24M seed round investment. The company is co-founded by LinkedIn founder, Reid Hoffman. Read more here.

🚀 C the Signs, the developer of AI-powered early cancer detection solutions is planning to expand beyond the United Kingdom to the US and has raised $8M towards this effort. We first covered C the Signs in the January roundup, highlighting its potential in colorectal cancer early detection. The platform has also shown strong promise in breast and ovarian cancer within within the UK’s National Health Service dataset. Read more here.

🚀 Oatmeal Health has launched ICARE, an AI-powered platform for lung cancer risk stratification, covered by Medicare. The platform “utilizes cutting-edge AI and computer vision to analyze 3D low-dose CT scans, automatically detecting, analyzing, and categorizing lung nodules based on their risk profile.” The AI platform is trained on a diverse dataset of over 100,000 CT scans and is designed to enhance accuracy and efficiency of lung cancer screening, particularly in underserved communities. Read more here.

🚀 Paige, a provider of innovative AI-powered cancer diagnostics technologies, announced the expansion of its groundbreaking multi-cancer detection tool called Paige PanCancer Detect. The tool, which originally launched in 2024 with a panel of 17 tissue types has expanded to 40 tissue and organ types. Paige’s PanCancer Suite is built on Virchow V2, one of the company’s foundation models trained on 3 million digitized slides. With this, the platform helps pathologists identify the most subtle tissue complexities across many more cancer types, including rare ones. Read more here.

🚀 Onc.ai announced that its prognostic tool, called Serial CTRS, that uses AI to categorize patients with non-small cell lung cancer (NSCLC) as high- or low-mortality risk, has been awarded Breakthrough Device Designation by the FDA. According to Akshay Nanduri, CEO of Onc.AI, the AI tool “aims to equip oncologists with vital, automated prognostic insights using routinely collected diagnostic imaging scans and ultimately improve treatment strategy and provide risk stratification throughout a cancer patient’s journey.” Read more here.

🚀 Certis Oncology announced US Patent issuance for its proprietary CertisAI™ predictive Oncology Intelligence Platform. CertisAI is the first and only commercially available AI/ML platform that predicts drug efficacy by identifying unique tumor gene expression patterns that correlate with drug response. This helps accelerate discovery and development efforts. Read more here.

🚀 Latent Labs, an AI-first biotech company focused on using generative AI to design therapeutic proteins, has emerged from stealth with a $40M series A funding. The company was founded by Simon Kohl, previously a co-lead of DeepMind’s protein design team and senior research scientist on DeepMind’s Nobel Prize-winning protein structure prediction tool - AlphaFold. Latent Labs is building AI foundation models to “make biology programmable,” and plans to work with biotech partners to generate and optimize proteins. Read more here.

🚀 Perthera, a precision oncology technology company, announced the launch of a new version of its Perthera Report^®, which uses a patient’s prior medical history and molecular test results to precisely rank therapy options. The new version, called Perthera Report^® Pancreatic version (PRPv), was purpose-designed for oncologists treating pancreatic cancer patients and incorporates Perthera’s recently announced frontline chemotherapy predictor, PDACai. We first covered Perthera’s PDACai in the January roundup. Read more here.