Siemens Healthineers AG Develops AI-Powered Computed Tomography System to Detect Pregnancy in Patients

Strategic Insights -

Siemens Healthineers AG has been issued a patent for an AI-powered computed tomography (CT) system designed to detect pregnancy in patients before undergoing a CT scan. The system involves using a trained machine learning model to analyze a two-dimensional radiograph of the patient and determine whether they are pregnant. If the patient is pregnant, the system can modify the CT scan to reduce radiation exposure to the fetus.

The inventors, Katharine Grant and Bernhard Schmidt, describe the need for a system that can efficiently detect patient conditions and adapt CT imaging parameters within a CT imaging workflow. They explain that conventional medical imaging systems can subject patients to significant radiation exposure, and that a system is desired to optimize the balance between diagnostic information and radiation exposure.

The patent includes several claims that outline the system's functionality, including planning a CT scan, acquiring a radiograph, inputting the radiograph to a trained machine learning model, and receiving a classification indicating whether the patient is pregnant or not. The system can modify the CT scan to reduce radiation exposure to reproductive organs if the patient is pregnant.

Key Takeaways:

  • Siemens Healthineers AG has developed an AI-powered CT system to detect pregnancy in patients.
  • The system uses a trained machine learning model to analyze a radiograph and determine whether the patient is pregnant.
  • If the patient is pregnant, the system can modify the CT scan to reduce radiation exposure to the fetus.
  • The system aims to optimize the balance between diagnostic information and radiation exposure.
  • The patent includes several claims that outline the system's functionality, including planning a CT scan, acquiring a radiograph, and modifying the CT scan based on the classification.
  • The system can be trained based on a plurality of radiographs and a classification corresponding to each radiograph.
  • The system can modify the CT scan to reduce radiation exposure to reproductive organs if the patient is pregnant.

Statistics:

  • Siemens Healthineers AG has developed an AI-powered CT system to detect pregnancy in patients.
  • The system uses a trained machine learning model to analyze a radiograph and determine whether the patient is pregnant.
  • 1.7 million patients are exposed to radiation from CT scans in the United States each year (Source: American College of Radiology).
  • Radiation exposure can increase the risk of cancer in patients (Source: World Health Organization).
  • The system aims to reduce radiation exposure to reproductive organs in pregnant patients.
  • The patent was filed on May 19, 2022, and published online on June 3, 2025.

Sources:

  • Grant, Katharine. Machine learning-assisted computed tomography workflow. U.S. Patent Number 12322090, filed May 19, 2022, and published online on June 3, 2025.
  • American College of Radiology. Radiation Exposure in CT Scans.
  • World Health Organization. Ionizing Radiation: Protecting People and the Environment.