Deep-Learning System for Automatic Measurement of the Femorotibial Rotational Angle on Lower-Extremity Computed Tomography.

J Imaging Inform Med

Department of Biomedical Engineering, Gachon University, Seongnam-Si 13120, Gyeonggi-Do, Republic of Korea.

Published: September 2025


Category Ranking

98%

Total Visits

921

Avg Visit Duration

2 minutes

Citations

20

Article Abstract

To develop and validate a deep-learning-based algorithm for automatic identification of anatomical landmarks and calculating femoral and tibial version angles (FTT angles) on lower-extremity CT scans. In this IRB-approved, retrospective study, lower-extremity CT scans from 270 adult patients (median age, 69 years; female to male ratio, 235:35) were analyzed. CT data were preprocessed using contrast-limited adaptive histogram equalization and RGB superposition to enhance tissue boundary distinction. The Attention U-Net model was trained using the gold standard of manual labeling and landmark drawing, enabling it to segment bones, detect landmarks, create lines, and automatically measure the femoral version and tibial torsion angles. The model's performance was validated against manual segmentations by a musculoskeletal radiologist using a test dataset. The segmentation model demonstrated 92.16%±0.02 sensitivity, 99.96%±<0.01 specificity, and 2.14±2.39 HD95, with a Dice similarity coefficient (DSC) of 93.12%±0.01. Automatic measurements of femoral and tibial torsion angles showed good correlation with radiologists' measurements, with correlation coefficients of 0.64 for femoral and 0.54 for tibial angles (p < 0.05). Automated segmentation significantly reduced the measurement time per leg compared to manual methods (57.5 ± 8.3 s vs. 79.6 ± 15.9 s, p < 0.05). We developed a method to automate the measurement of femorotibial rotation in continuous axial CT scans of patients with osteoarthritis (OA) using a deep-learning approach. This method has the potential to expedite the analysis of patient data in busy clinical settings.

Download full-text PDF

Source
http://dx.doi.org/10.1007/s10278-025-01641-0DOI Listing

Publication Analysis

Top Keywords

lower-extremity scans
8
deep-learning system
4
system automatic
4
automatic measurement
4
measurement femorotibial
4
femorotibial rotational
4
rotational angle
4
angle lower-extremity
4
lower-extremity computed
4
computed tomography
4

Similar Publications

To develop and validate a deep-learning-based algorithm for automatic identification of anatomical landmarks and calculating femoral and tibial version angles (FTT angles) on lower-extremity CT scans. In this IRB-approved, retrospective study, lower-extremity CT scans from 270 adult patients (median age, 69 years; female to male ratio, 235:35) were analyzed. CT data were preprocessed using contrast-limited adaptive histogram equalization and RGB superposition to enhance tissue boundary distinction.

View Article and Find Full Text PDF

Achieving optimal alignment and fit is a key aspect of ankle-foot orthosis (AFO) design, as it directly influences the effectiveness of the device. While digital workflows offer the potential to integrate quantifiable alignment measures and corrections into AFO design, a major challenge remains in controlling lower-limb positioning and alignment during 3D scanning. This study aimed to evaluate pediatric AFO alignment and shape differences of directly scanned (live scan) vs casted lower limb models.

View Article and Find Full Text PDF

A 37-year-old man presented with swelling and erythema in the left first toe after a prior trauma, suspicious for osteomyelitis. X-ray and computed tomography (CT) scans revealed a radiolucent lesion with cortical disruption. A 99mTc/tricine/HYNIC ubiquicidin 29-41 (UBI) scintigraphy showed increased uptake but a non-accumulative time-activity curve, indicating a false positive for infection.

View Article and Find Full Text PDF

Objective: Multiple studies have demonstrated an increased risk of lower extremity injuries for females in frontal crashes. This study aimed to investigate whether sex-based anatomical differences, as measured on computed tomography (CT) scans of the abdomen and pelvis, contribute to lower extremity injury risk.

Methods: The Crash Injury Research and Engineering Network (CIREN) database (2017-2023) was queried for frontal collisions.

View Article and Find Full Text PDF

Experimental Evaluation of Maximum-Likelihood-Based Data Preconditioning for DE-SPECT: A Clinical SPECT System Constructed With CZT Imaging Detectors.

IEEE Trans Radiat Plasma Med Sci

May 2025

Department of Nuclear, Plasma and Radiological Engineering, and the Department of Bioengineering and Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana-Champaign, Champaign, IL 61801 USA.

This study introduces a novel maximum-likelihood-based data preconditioning method for a 3-D position sensitive cadmium zinc telluride (CZT) detector used in the dynamic extremity-single photon emission computed tomography imaging system, an organ-dedicated Single-Photon Emission computed tomography system optimized for imaging peripheral vascular diseases in lower extremities. The 3-D CZT detectors offer subpixel resolution of ~0.5 mm FWHM in directions and an ultrahigh energy resolution of 3 keV at 200 keV, 4.

View Article and Find Full Text PDF