28 November - 1 Dicember

RSNA 2021, join us!

News

Bone Suppression

18/11/2021

Bone suppression uses Artificial Intelligence to automatically remove bones from standard chest X-ray image with a single exposure, increasing the quality of soft tissue visibility.

Chest radiographs are the most commonly used imaging technique for diagnosis of lung diseases and abnormalities, due to their efficiency, low radiation dose, low cost and their portability.

However, the presence of osseous structures such as ribs and clavicles can partially or completely obscure lung lesions, resulting in diagnostic errors and missed diagnosis. Bone suppression technique addresses this issue offering a clearer examination of the lung field by reducing the noise caused by high-contrast bone structures on the chest radiograph.

The technology uses existing chest x-ray images without any special equipment or increase in radiation dose.

AI-based approach

Digitec Bone suppression is an innovative AI solution based on a deep neural network properly trained to recognize the bone structures and remove them from a conventional chest x-ray image.

The input data is the acquired radiograph which is processed through AI model to accurately remove bone components. The output obtained is the corresponding bone-free image, similar to that obtained from dual-energy procedure without the need of two exposure.

After the acquisition of a standard chest x-ray image the Bone suppression algorithm is applied in the background and both images are automatically available at any time. 

Clinical Impact and other advantages

From a clinical point of view the interpretation of chest radiographs is very challenging. Thus, an increased sensitivity for the detection of lung abnormalities represents a desirable diagnostic improvement.

A better diagnostic performance is provided by a better visualization of the lung field and thus, lesions otherwise obscured from the bone structures. Existing literature reported the risk of missed lung cancers in the chest radiography due to the difficulty to distinguish a lung lesion from bones.

Bone suppressed images have been found to significantly improve the radiologists’ diagnostic performance and confidence for detection of pulmonary abnormalities such as lung nodules, lung cancer, metastasis, pneumothorax, pneumonia, and other pathologic lung changes.

The main benefits are reported:

  • Improving efficiency of chest radiograph interpretation by increasing the quality of soft-tissues visibility. 
  • Improving the detection of small and subtle lung lesions.
  • Supporting radiologists during the decision-making process.
  • Reducing interpretation errors.
  • Support the prevention of overlooking diseases.
  • Software-based application with no special equipment needed.
  • No additional x-ray dose or examinations times are required.
  • Easy-of-use and portability.
  • No impact on the standard workflow.

Moreover, it has been proved that the experience level of clinicians has a great impact on diagnostic performance: bone suppression could reduce the impact, helping less experience radiologists in the detection of lung lesions.

Summary

Digitec Bone suppression is able to enhance the value of screening digital chest radiography, combining the chest x-ray practicality with the high quality of an AI image processing, in a manner that is efficient for both clinicians and patients.

References

[1] Hong GS, Do KH, Lee CW. Added Value of Bone Suppression Image in the Detection of Subtle Lung Lesions on Chest Radiographs with Regard to Reader’s Expertise. J Korean Med Sci. 2019 Oct 7;34(38):e250. doi: 10.3346/jkms.2019.34.e250. PMID: 31583870; PMCID: PMC6776835.

[2] Li F, Engelmann R, Pesce LL, Doi K, Metz CE, Macmahon H. Small lung cancers: improved detection by use of bone suppression imaging—comparison with dual-energy subtraction chest radiography. Radiology 2011; 261(3):937–949.