Opa:
Patologian AI-polku on näköjään myös Alibaban kartalla, haimasyövän AI-mallia USA:n suuntaan.
Hieman mietityttää, että mitenhän regulaatio-/ hiturimaailmassa pärjätään kilpailussa alan menestyksessä…
Investing.com – Alibaba (NYSE:BABA ) Group’s research division Damo Academy has received a “breakthrough device” designation from the U.S. Food and Drug Administration (FDA) for its pancreatic cancer-detecting AI tool, according to the South China Morning Post. The designation allows for an expedited path toward regulatory approval in the United States.
The AI model, called Damo Panda, was designed to detect early-stage pancreatic cancer using deep learning and abdominal CT scans. According to researchers, the model demonstrated significantly higher sensitivity than human radiologists in initial tests.
Damo Panda was first introduced in November 2023 in a study published in Nature Medicine. The paper outlined that the tool was trained on data from over 3,200 patients and achieved 34.1% higher sensitivity in detecting pancreatic malignancies.
Pancreatic cancer is notoriously difficult to detect early, and early diagnosis can significantly improve patient survival rates. Alibaba’s tool aims to scan asymptomatic patients for early signs of the disease, potentially improving medical outcomes through earlier intervention.
Alibaba has already conducted clinical trials with the model in China, SCMP reported, citing a company statement released Thursday. The company said the breakthrough designation in the U.S. could accelerate efforts to bring Damo Panda to global healthcare markets.
The FDA’s program is intended to speed up review processes for devices that offer more effective treatment or diagnosis for life-threatening conditions. The designation does not equate to commercial approval but can reduce regulatory hurdles for emerging technologies.
The move also underscores China’s expanding role in AI-driven medical innovation. Firms like Alibaba are increasingly investing in life sciences and digital health as part of broader diversification strategies.
Tuo ei ole patologiaa, vaan kuvantamista eli radiologiaa. Analysoidaan tietokonekerroskuvia eli CT-skannausta. Ei kilpaile Aiforian kanssa. Sinänsä Aiforian mallejakin voidaan kouluttaa tekemään mitä tahansa kuvantunnistusta, mutta tuonne ei olla menossa. Kiina tulee kyllä kaikilla rintamilla DeepSeekin tapaan.
20 tykkäystä