دورية أكاديمية
THA-Net: A Deep Learning Solution for Next-Generation Templating and Patient-specific Surgical Execution.
| العنوان: | THA-Net: A Deep Learning Solution for Next-Generation Templating and Patient-specific Surgical Execution. |
|---|---|
| المؤلفون: | Rouzrokh P; Department of Radiology, Mayo Clinic, Minnesota., Khosravi B; Department of Radiology, Mayo Clinic, Minnesota., Mickley JP; Department of Orthopedic Surgery, Mayo Clinic, Minnesota., Erickson BJ; Department of Radiology, Mayo Clinic, Minnesota., Taunton MJ; Department of Orthopedic Surgery, Mayo Clinic, Minnesota., Wyles CC; Department of Orthopedic Surgery, Mayo Clinic, Minnesota. |
| المصدر: | The Journal of arthroplasty [J Arthroplasty] 2024 Mar; Vol. 39 (3), pp. 727-733.e4. Date of Electronic Publication: 2023 Aug 22. |
| نوع المنشور: | Journal Article |
| اللغة: | English |
| بيانات الدورية: | Publisher: Taylor and Francis Country of Publication: United States NLM ID: 8703515 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1532-8406 (Electronic) Linking ISSN: 08835403 NLM ISO Abbreviation: J Arthroplasty Subsets: MEDLINE |
| أسماء مطبوعة: | Publication: New Brunswick, NJ : Taylor and Francis Original Publication: [New York, NY : Churchill Livingstone, c1986- |
| مواضيع طبية MeSH: | Arthroplasty, Replacement, Hip*/methods , Hip Prosthesis* , Deep Learning*, Humans ; Hip Joint/diagnostic imaging ; Hip Joint/surgery ; Radiography ; Retrospective Studies |
| مستخلص: | Background: This study introduces THA-Net, a deep learning inpainting algorithm for simulating postoperative total hip arthroplasty (THA) radiographs from a single preoperative pelvis radiograph input, while being able to generate predictions either unconditionally (algorithm chooses implants) or conditionally (surgeon chooses implants). Methods: The THA-Net is a deep learning algorithm which receives an input preoperative radiograph and subsequently replaces the target hip joint with THA implants to generate a synthetic yet realistic postoperative radiograph. We trained THA-Net on 356,305 pairs of radiographs from 14,357 patients from a single institution's total joint registry and evaluated the validity (quality of surgical execution) and realism (ability to differentiate real and synthetic radiographs) of its outputs against both human-based and software-based criteria. Results: The surgical validity of synthetic postoperative radiographs was significantly higher than their real counterparts (mean difference: 0.8 to 1.1 points on 10-point Likert scale, P < .001), but they were not able to be differentiated in terms of realism in blinded expert review. Synthetic images showed excellent validity and realism when analyzed with already validated deep learning models. Conclusion: We developed a THA next-generation templating tool that can generate synthetic radiographs graded higher on ultimate surgical execution than real radiographs from training data. Further refinement of this tool may potentiate patient-specific surgical planning and enable technologies such as robotics, navigation, and augmented reality (an online demo of THA-Net is available at: https://demo.osail.ai/tha_net). (Copyright © 2023 Elsevier Inc. All rights reserved.) |
| فهرسة مساهمة: | Keywords: artificial intelligence; deep learning; denoising diffusion probabilistic models; machine learning; templating; total hip arthroplasty |
| تواريخ الأحداث: | Date Created: 20230824 Date Completed: 20240214 Latest Revision: 20240214 |
| رمز التحديث: | 20240214 |
| DOI: | 10.1016/j.arth.2023.08.063 |
| PMID: | 37619804 |
| قاعدة البيانات: | MEDLINE |
Full Text via ClinicalKey 
Full Text Finder | تدمد: | 1532-8406 |
|---|---|
| DOI: | 10.1016/j.arth.2023.08.063 |