Deep Learning-based calculation of patient size and attenuation surrogates from localizer Image: Toward personalized chest CT protocol optimization. Issue 157 (December 2022)

• Record Type: Journal Article
• Authors: Salimi, Yazdan; Shiri, Isaac; Akhavanallaf, Azadeh; Mansouri, Zahra; Sanaat, AmirHosein; Pakbin, Masoumeh; Ghasemian, Mohammadreza; Arabi, Hossein; Zaidi, Habib
• Is Part Of: European journal of radiology. Issue 157(2022)
• Publication Date: 2022-12
• Subjects: X-ray computed tomography -- Radiation dose -- Deep learning -- Phantoms -- Body size; CT Computed tomography -- CF Conversion Factor -- AP Anterior-Posterior -- PA Posterior-Anterior -- DW water equivalent diameter -- ED Effective Dose -- TCM Tube Current Modulation -- DL Deep Learning -- CTDIvol Volumetric CT Dose Index -- Deff Effective Diameter -- SSDE Size-Specific Dose Estimate -- DLP Dose Length Product -- FOV Field-of-View -- TH Table Height -- GT Ground truth -- RL Right to Left -- CC Cranio-caudal -- SSIM Structural Similarity Index measure -- RAE Relative Absolute Error -- MAE Mean Absolute Error -- BMI Body Mass Index; Medical radiology -- Periodicals; Radiology -- Periodicals; Radiologie médicale -- Périodiques; Medical radiology; Periodicals; 616.075705
• DOI: 10.1016/j.ejrad.2022.110602
• Languages: English
• ISSNs: 0720-048X

Abstract

Highlights: Patient size measurement is required for CT patient dose evaluation and acquisition protocol optimization. Our proposed deep learning-based algorithm can measure a patient's body size, attenuation and shape parameters of the body area, water equivalent diameter, and body contour with an error of<6%. Our proposed method can generate 3D axial 64-slice CT images (output) from a single 2D localizer (input) prior to the main spiral CT scan. Abstract: Purpose: Extracting water equivalent diameter (DW), as a good descriptor of patient size, from the CT localizer before the spiral scan not only minimizes truncation errors due to the limited scan field-of-view but also enables prior size-specific dose estimation as well as scan protocol optimization. This study proposed a unified methodology to measure patient size, shape, and attenuation parameters from a 2D anterior-posterior localizer image using deep learning algorithms without the need for labor-intensive vendor-specific calibration procedures. Methods: 3D CT chest images and 2D localizers were collected for 4005 patients. A modified U-NET architecture was trained to predict the 3D CT images from their corresponding localizer scans. The algorithm was tested on 648 and 138 external cases with fixed and variable table height positions. To evaluate the performance of the prediction model, structural similarity index measure (SSIM), body area, body contour, Dice index, and water equivalent diameter (DW) were calculated and compared between the predicted 3D CT images and the ground truth (GT) images in a slicewise manner. Results: The average age of the patients included in this study (1827 male and 1554 female) was 53.8 ± 17.9 (18–120) years. The DW, tube current, and CTDIvol measured on original axial images in the external 138 cases group were significantly larger than those of the external 648 cases (P < 0.05). The SSIM and Dice index calculated between the prediction and GT for body contour were 0.998 ± 0.001 and 0.950 ± 0.016, respectively. The average percentage error in the calculation of DW was 2.7 ± 3.5 %. The error in the DW calculation was more considerable in larger patients (p-value < 0.05). Conclusions: We developed a model to predict the patient size, shape, and attenuation factors slice-by-slice prior to spiral scanning. The model exhibited remarkable robustness to table height variations. The estimated parameters are helpful for patient dose reduction and protocol optimization.