A CT-based Deep Learning Model for Predicting Subsequent Fracture Risk in Patients with Hip Fracture.

Radiology

From the Interdisciplinary Program in Bioengineering (Y.K., Y.S.) and Integrated Major in Innovative Medical Science (Y.K.), Seoul National University Graduate School, Seoul, Republic of Korea; Department of Radiology (Y.K.), Transdisciplinary Department of Medicine & Advanced Technology (Y.G.K., B.W.K., Y.S.), and Department of Internal Medicine (J.H.K., C.S.S.), Seoul National University Hospital, Seoul, Republic of Korea; Departments of Orthopaedic Surgery (J.W.P., Y.K.L.) and Internal Medicine (S.H.K.), Seoul National University Bundang Hospital, 82 Gumi-ro 173 Beon-gil, Bundang gu, Seongnam, Republic of Korea; Departments of Medicine (Y.G.K.) and Internal Medicine (S.H.K., J.H.K., S.W.K., C.S.S.), Seoul National University College of Medicine, Seoul, Republic of Korea; and Department of Internal Medicine, Seoul National University Boramae Hospital, Seoul, Republic of Korea (S.W.K.).

Published: January 2024

AI Article Synopsis

  • A study aimed to create a deep learning model to predict the risk of subsequent fractures in patients who recently experienced a hip fracture, by using digitally reconstructed radiographs from hip CT scans.
  • The research analyzed data from 1,012 patients for model development and validated it on another 468 patients, finding that the ensemble model significantly outperformed other existing prediction models in terms of accuracy over 2, 3, and 5 years.
  • Results indicated that the new model had higher probabilities of predicting fractures compared to both image-based models and a clinical model that considered established risk factors, showcasing its potential for improving patient care.

Article Abstract

Background Patients have the highest risk of subsequent fractures in the first few years after an initial fracture, yet models to predict short-term subsequent risk have not been developed. Purpose To develop and validate a deep learning prediction model for subsequent fracture risk using digitally reconstructed radiographs from hip CT in patients with recent hip fractures. Materials and Methods This retrospective study included adult patients who underwent three-dimensional hip CT due to a fracture from January 2004 to December 2020. Two-dimensional frontal, lateral, and axial digitally reconstructed radiographs were generated and assembled to construct an ensemble model. DenseNet modules were used to calculate risk probability based on extracted image features and fracture-free probability plots were output. Model performance was assessed using the C index and area under the receiver operating characteristic curve (AUC) and compared with other models using the paired test. Results The training and validation set included 1012 patients (mean age, 74.5 years ± 13.3 [SD]; 706 female, 113 subsequent fracture) and the test set included 468 patients (mean age, 75.9 years ± 14.0; 335 female, 22 subsequent fractures). In the test set, the ensemble model had a higher C index (0.73) for predicting subsequent fractures than that of other image-based models (C index range, 0.59-0.70 for five of six models; value range, < .001 to < .05). The ensemble model achieved AUCs of 0.74, 0.74, and 0.73 at the 2-, 3-, and 5-year follow-ups, respectively; higher than that of most other image-based models at 2 years (AUC range, 0.57-0.71 for five of six models; value range, < .001 to < .05) and 3 years (AUC range, 0.55-0.72 for four of six models; value range, < .001 to < .05). Moreover, the AUCs achieved by the ensemble model were higher than that of a clinical model that included known risk factors (2-, 3-, and 5-year AUCs of 0.58, 0.64, and 0.70, respectively; < .001 for all). Conclusion In patients with recent hip fractures, the ensemble deep learning model using digital reconstructed radiographs from hip CT showed good performance for predicting subsequent fractures in the short term. © RSNA, 2024 See also the editorial by Li and Jaremko in this issue.

Download full-text PDF

Source
http://dx.doi.org/10.1148/radiol.230614DOI Listing

Publication Analysis

Top Keywords

subsequent fractures
16
ensemble model
16
models range
16
deep learning
12
predicting subsequent
12
subsequent fracture
12
patients hip
12
reconstructed radiographs
12
range 001
12
model
9

Similar Publications

Background: Cementoplasty has been successfully used for treating fractures in various parts of the human body, although the use in weight-bearing long bones is a subject of controversial debate. Strategies to improve the mechanical properties of polymethylmethacrylate-based bone cement (BC) comprise changing the chemical composition or the application of metal reinforcement strategies. In clinical practice reinforced bone cement is used despite biomechanical basic research regarding this topic being scare.

View Article and Find Full Text PDF

Background: Hip fractures are a source of severe pain among the elderly population and pose challenges due to limited analgesic tolerance. Perioperative methadone has shown promise in our pilot study suggesting a safe dose of 0.10 mg/kg, prompting further investigation into its benefits for elderly hip fracture patients.

View Article and Find Full Text PDF

Understanding he impact of dry-wet and freeze-thaw cycles on the mechanical properties of unloaded damaged rock masses in reservoir bank slopes is crucial for revealing the deformation and failure mechanisms in artificially excavated slope rock masses within fluctuation zones. To address, the study focuses on unloaded damaged samples subjected to excavation disturbances, conducting various cycles of dry-wet and freeze-thaw treatment along with uniaxial and triaxial re-loading tests. A damage statistical constitutive model was established based on the experimental results and validated using numerical simulation methods.

View Article and Find Full Text PDF

Paradoxical Agitation and Masseter Spasm During Propofol Procedural Sedation: A Case Report.

Clin Pract Cases Emerg Med

November 2024

Christiana Care Health System, Department of Emergency Medicine, Newark, Delaware.

Introduction: Propofol is an anesthetic agent commonly used in emergency department (ED) procedural sedation. It is often preferred in orthopedic procedures because of its muscle-relaxing properties. Rarely, however, it can induce agitation and muscle hypertonicity.

View Article and Find Full Text PDF

Objective: This study was aimed at assessing the effects of various finish line designs and cement gap thicknesses on the fracture resistance of gradient zirconia crowns.

Methods: Sixty crowns were fabricated on stainless-steel dies with yttria multi-layered (YML) zirconia and categorized into three primary groups according to finish line type (knife-edge, chamfer, and rounded shoulder). Each group was further classified into two subgroups (n = 10) according to cement space thickness (50 or 80 μm).

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!