Context-Aware Dual-Task Deep Network for Concurrent Bone Segmentation and Clinical Assessment to Enhance Shoulder Arthroplasty Preoperative planning
<italic>Goal:</italic> Effective preoperative planning for shoulder joint replacement requires accurate glenohumeral joint (GH) digital surfaces and reliable clinical staging. <italic>Methods:</italic> xCEL-UNet was designed as a dual-task deep network for humerus and scapula...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Open Journal of Engineering in Medicine and Biology |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10835174/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832583988886110208 |
|---|---|
| author | Luca Marsilio Andrea Moglia Alfonso Manzotti Pietro Cerveri |
| author_facet | Luca Marsilio Andrea Moglia Alfonso Manzotti Pietro Cerveri |
| author_sort | Luca Marsilio |
| collection | DOAJ |
| description | <italic>Goal:</italic> Effective preoperative planning for shoulder joint replacement requires accurate glenohumeral joint (GH) digital surfaces and reliable clinical staging. <italic>Methods:</italic> xCEL-UNet was designed as a dual-task deep network for humerus and scapula bone reconstruction in CT scans, and assessment of three GH joint clinical conditions, namely osteophyte size (OS), joint space reduction (JS), and humeroscapular alignment (HSA). <italic>Results:</italic> Trained on a dataset of 571 patients, the model optimized segmentation and classification through transfer learning. It achieved median root mean squared errors of 0.31 and 0.24 mm, and Hausdorff distances of 2.35 and 3.28 mm for the humerus and scapula, respectively. Classification accuracy was 91 for OS, 93 for JS, and 85% for HSA. GradCAM-based activation maps validated the network's interpretability. <italic>Conclusions:</italic> this framework delivers accurate 3D bone surface reconstructions and dependable clinical assessments of the GH joint, offering robust support for therapeutic decision-making in shoulder arthroplasty. |
| format | Article |
| id | doaj-art-8f27caacca814701baebc82b5351c9bf |
| institution | Kabale University |
| issn | 2644-1276 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Open Journal of Engineering in Medicine and Biology |
| spelling | doaj-art-8f27caacca814701baebc82b5351c9bf2025-01-28T00:02:11ZengIEEEIEEE Open Journal of Engineering in Medicine and Biology2644-12762025-01-01626927810.1109/OJEMB.2025.352787710835174Context-Aware Dual-Task Deep Network for Concurrent Bone Segmentation and Clinical Assessment to Enhance Shoulder Arthroplasty Preoperative planningLuca Marsilio0https://orcid.org/0009-0001-9738-0438Andrea Moglia1https://orcid.org/0000-0002-3365-580XAlfonso Manzotti2https://orcid.org/0000-0003-1791-6800Pietro Cerveri3https://orcid.org/0000-0003-3995-8673Department of Electronics, Information and Bioengineering, Politecnico di Milano, Milan, ItalyDepartment of Electronics, Information and Bioengineering, Politecnico di Milano, Milan, ItalyHospital ASST FBF-Sacco, Milan, ItalyDepartment of Electronics, Information and Bioengineering, Politecnico di Milano, Milan, Italy<italic>Goal:</italic> Effective preoperative planning for shoulder joint replacement requires accurate glenohumeral joint (GH) digital surfaces and reliable clinical staging. <italic>Methods:</italic> xCEL-UNet was designed as a dual-task deep network for humerus and scapula bone reconstruction in CT scans, and assessment of three GH joint clinical conditions, namely osteophyte size (OS), joint space reduction (JS), and humeroscapular alignment (HSA). <italic>Results:</italic> Trained on a dataset of 571 patients, the model optimized segmentation and classification through transfer learning. It achieved median root mean squared errors of 0.31 and 0.24 mm, and Hausdorff distances of 2.35 and 3.28 mm for the humerus and scapula, respectively. Classification accuracy was 91 for OS, 93 for JS, and 85% for HSA. GradCAM-based activation maps validated the network's interpretability. <italic>Conclusions:</italic> this framework delivers accurate 3D bone surface reconstructions and dependable clinical assessments of the GH joint, offering robust support for therapeutic decision-making in shoulder arthroplasty.https://ieeexplore.ieee.org/document/10835174/Assisted preoperative planningdeep learningexplainable AIshoulder arthroplastyshoulder bone segmentation |
| spellingShingle | Luca Marsilio Andrea Moglia Alfonso Manzotti Pietro Cerveri Context-Aware Dual-Task Deep Network for Concurrent Bone Segmentation and Clinical Assessment to Enhance Shoulder Arthroplasty Preoperative planning IEEE Open Journal of Engineering in Medicine and Biology Assisted preoperative planning deep learning explainable AI shoulder arthroplasty shoulder bone segmentation |
| title | Context-Aware Dual-Task Deep Network for Concurrent Bone Segmentation and Clinical Assessment to Enhance Shoulder Arthroplasty Preoperative planning |
| title_full | Context-Aware Dual-Task Deep Network for Concurrent Bone Segmentation and Clinical Assessment to Enhance Shoulder Arthroplasty Preoperative planning |
| title_fullStr | Context-Aware Dual-Task Deep Network for Concurrent Bone Segmentation and Clinical Assessment to Enhance Shoulder Arthroplasty Preoperative planning |
| title_full_unstemmed | Context-Aware Dual-Task Deep Network for Concurrent Bone Segmentation and Clinical Assessment to Enhance Shoulder Arthroplasty Preoperative planning |
| title_short | Context-Aware Dual-Task Deep Network for Concurrent Bone Segmentation and Clinical Assessment to Enhance Shoulder Arthroplasty Preoperative planning |
| title_sort | context aware dual task deep network for concurrent bone segmentation and clinical assessment to enhance shoulder arthroplasty preoperative planning |
| topic | Assisted preoperative planning deep learning explainable AI shoulder arthroplasty shoulder bone segmentation |
| url | https://ieeexplore.ieee.org/document/10835174/ |
| work_keys_str_mv | AT lucamarsilio contextawaredualtaskdeepnetworkforconcurrentbonesegmentationandclinicalassessmenttoenhanceshoulderarthroplastypreoperativeplanning AT andreamoglia contextawaredualtaskdeepnetworkforconcurrentbonesegmentationandclinicalassessmenttoenhanceshoulderarthroplastypreoperativeplanning AT alfonsomanzotti contextawaredualtaskdeepnetworkforconcurrentbonesegmentationandclinicalassessmenttoenhanceshoulderarthroplastypreoperativeplanning AT pietrocerveri contextawaredualtaskdeepnetworkforconcurrentbonesegmentationandclinicalassessmenttoenhanceshoulderarthroplastypreoperativeplanning |