1
|
Wernér K, Anttila T, Hulkkonen S, Viljakka T, Haapamäki V, Ryhänen J. Detecting Avascular Necrosis of the Lunate from Radiographs Using a Deep-Learning Model. J Imaging Inform Med 2024; 37:706-714. [PMID: 38343256 PMCID: PMC11031541 DOI: 10.1007/s10278-023-00964-0] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 04/23/2023] [Revised: 11/04/2023] [Accepted: 11/07/2023] [Indexed: 04/20/2024]
Abstract
Deep-learning (DL) algorithms have the potential to change medical image classification and diagnostics in the coming decade. Delayed diagnosis and treatment of avascular necrosis (AVN) of the lunate may have a detrimental effect on patient hand function. The aim of this study was to use a segmentation-based DL model to diagnose AVN of the lunate from wrist postero-anterior radiographs. A total of 319 radiographs of the diseased lunate and 1228 control radiographs were gathered from Helsinki University Central Hospital database. Of these, 10% were separated to form a test set for model validation. MRI confirmed the absence of disease. In cases of AVN of the lunate, a hand surgeon at Helsinki University Hospital validated the accurate diagnosis using either MRI or radiography. For detection of AVN, the model had a sensitivity of 93.33% (95% confidence interval (CI) 77.93-99.18%), specificity of 93.28% (95% CI 87.18-97.05%), and accuracy of 93.28% (95% CI 87.99-96.73%). The area under the receiver operating characteristic curve was 0.94 (95% CI 0.88-0.99). Compared to three clinical experts, the DL model had better AUC than one clinical expert and only one expert had higher accuracy than the DL model. The results were otherwise similar between the model and clinical experts. Our DL model performed well and may be a future beneficial tool for screening of AVN of the lunate.
Collapse
Affiliation(s)
- Krista Wernér
- Musculoskeletal and Plastic Surgery, Department of Hand Surgery, University of Helsinki and Helsinki University Hospital, Haartmaninkatu 4, 00290, Helsinki, Finland.
| | - Turkka Anttila
- Musculoskeletal and Plastic Surgery, Department of Hand Surgery, University of Helsinki and Helsinki University Hospital, Haartmaninkatu 4, 00290, Helsinki, Finland
| | - Sina Hulkkonen
- Musculoskeletal and Plastic Surgery, Department of Hand Surgery, University of Helsinki and Helsinki University Hospital, Haartmaninkatu 4, 00290, Helsinki, Finland
| | | | - Ville Haapamäki
- Department of Radiology, HUS Diagnostic Center, HUS Medical Imaging Center, Helsinki, Finland
| | - Jorma Ryhänen
- Musculoskeletal and Plastic Surgery, Department of Hand Surgery, University of Helsinki and Helsinki University Hospital, Haartmaninkatu 4, 00290, Helsinki, Finland
| |
Collapse
|
2
|
Anttila TT, Aspinen S, Pierides G, Haapamäki V, Laitinen MK, Ryhänen J. Enchondroma Detection from Hand Radiographs with an Interactive Deep Learning Segmentation Tool-A Feasibility Study. J Clin Med 2023; 12:7129. [PMID: 38002741 PMCID: PMC10672653 DOI: 10.3390/jcm12227129] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/24/2023] [Revised: 11/01/2023] [Accepted: 11/13/2023] [Indexed: 11/26/2023] Open
Abstract
Enchondromas are common benign bone tumors, usually presenting in the hand. They can cause symptoms such as swelling and pain but often go un-noticed. If the tumor expands, it can diminish the bone cortices and predispose the bone to fracture. Diagnosis is based on clinical investigation and radiographic imaging. Despite their typical appearance on radiographs, they can primarily be misdiagnosed or go totally unrecognized in the acute trauma setting. Earlier applications of deep learning models to image classification and pattern recognition suggest that this technique may also be utilized in detecting enchondroma in hand radiographs. We trained a deep learning model with 414 enchondroma radiographs to detect enchondroma from hand radiographs. A separate test set of 131 radiographs (47% with an enchondroma) was used to assess the performance of the trained deep learning model. Enchondroma annotation by three clinical experts served as our ground truth in assessing the deep learning model's performance. Our deep learning model detected 56 enchondromas from the 62 enchondroma radiographs. The area under receiver operator curve was 0.95. The F1 score for area statistical overlapping was 69.5%. Our deep learning model may be a useful tool for radiograph screening and raising suspicion of enchondroma.
Collapse
Affiliation(s)
- Turkka Tapio Anttila
- Musculoskeletal and Plastic Surgery, Department of Hand Surgery, University of Helsinki and Helsinki University Hospital, 00029 Helsinki, Finland
| | - Samuli Aspinen
- Musculoskeletal and Plastic Surgery, Department of Hand Surgery, University of Helsinki and Helsinki University Hospital, 00029 Helsinki, Finland
| | - Georgios Pierides
- Musculoskeletal and Plastic Surgery, Department of Hand Surgery, University of Helsinki and Helsinki University Hospital, 00029 Helsinki, Finland
| | - Ville Haapamäki
- Department of Radiology, University of Helsinki and Helsinki University Hospital, 00029 Helsinki, Finland
| | - Minna Katariina Laitinen
- Musculoskeletal and Plastic Surgery, Department of Orthopedic Surgery, University of Helsinki and Helsinki University Hospital, 00029 Helsinki, Finland
| | - Jorma Ryhänen
- Musculoskeletal and Plastic Surgery, Department of Hand Surgery, University of Helsinki and Helsinki University Hospital, 00029 Helsinki, Finland
| |
Collapse
|
3
|
Mattila S, Waris E. Outcomes of Revision of Interposition Implant Trapeziometacarpal Arthroplasty. Hand (N Y) 2023; 18:57S-64S. [PMID: 34301157 PMCID: PMC10052631 DOI: 10.1177/15589447211028920] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 11/16/2022]
Abstract
BACKGROUND Implant arthroplasties for trapeziometacarpal osteoarthritis are often associated with high complication and revision surgery rates. There are no previous studies reporting revision outcomes of failed interposition implant arthroplasty. METHODS A patient register search was done for all implant arthroplasties performed for trapeziometacarpal osteoarthritis during a 10-year period in a single hand surgical unit. Altogether, 32 patients had primary interposition implant arthroplasty (Artelon 22, Pyrosphere 6, Ortosphere 2, and Pyrodisk 2), and 19 of these patients had revision surgery with 23 revision procedures performed. In all, 15 of the revised 19 patients were reexamined clinically (Connolly-Rath score, Quick Disabilities of the Arm Shoulder and Hand, patient evaluation measure, the visual analog score for pain, thumb range of motion and strength measurements) and radiographically. RESULTS The indication for revision surgery was pain alone or implant dislocation accompanied by pain in all cases. Thirteen of the revised 15 patients reported functional deficit and pain after revision. There was no statistically significant difference in the revision outcomes between patients operated on primarily with the Artelon implant versus pyrocarbon/ceramic implants. Compared to previous studies on revision surgery and primary trapeziometacarpal arthroplasty, our results showed slightly higher pain and poorer functional scores. CONCLUSIONS Interposition implant arthroplasty may yield high revision rates. The results after revision surgery may be worse than previously described, and there may also be a tendency for worse results than those of primary arthroplasty. Interposition implant arthroplasty should always be thoroughly contemplated.
Collapse
Affiliation(s)
- Simo Mattila
- Helsinki University Central Hospital and University of Helsinki, Finland
| | - Eero Waris
- Helsinki University Central Hospital and University of Helsinki, Finland
| |
Collapse
|
4
|
Suominen EN, Sajanti AJ, Silver EA, Koivunen V, Bondfolk AS, Koskimäki J, Saarinen AJ. Alcohol intoxication and lack of helmet use are common in electric scooter-related traumatic brain injuries: a consecutive patient series from a tertiary university hospital. Acta Neurochir (Wien) 2022; 164:643-653. [PMID: 35029763 PMCID: PMC8759433 DOI: 10.1007/s00701-021-05098-2] [Citation(s) in RCA: 4] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/18/2021] [Accepted: 12/19/2021] [Indexed: 11/27/2022]
Abstract
PURPOSE Clinicians have increasingly encountered traumatic brain injuries (TBI) related to electric scooter (ES) accidents. In this study, we aim to identify the modifiable risk factors for ES-related TBIs. METHODS A retrospective cohort of consecutive patients treated for ES-related traumatic brain injuries in a tertiary university hospital between May 2019 and September 2021 was identified and employed for the study. The characteristics of the accidents along with the clinical and imaging findings of the injuries were collected from the patient charts. RESULTS During the study period, 104 TBIs related to ES accidents were identified. There was a high occurrence of accidents late at night and on Saturdays. In four cases, the patient's helmet use was mentioned (3.8%). Seventy-four patients (71%) were intoxicated. At the scene of the accident, seventy-seven (74%) of the patients had a Glasgow Coma Scale score of 13-15, three patients (3%) had a score of 9-12, and two patients (2%) had a score of 3-8. The majority (83%) of TBIs were diagnosed as concussions. Eighteen patients had evidence of intracranial injuries in the imagining. Two patients required neurosurgical procedures. The estimated population standardized incidence increased from 7.0/100,000 (95% CI 3.5-11/100,000) in 2019 to 27/100,000 (95% CI 20-34/100,000) in 2021. CONCLUSIONS Alcohol intoxication and the lack of a helmet were common in TBIs caused by ES accidents. Most of the accidents occurred late at night. Targeting these modifiable factors could decrease the incidence of ES-related TBIs.
Collapse
Affiliation(s)
- Eetu N Suominen
- Department of Paediatric Orthopaedic Surgery, University of Turku and Turku University Hospital, Kiinamyllynkatu 4-8, 20521, Turku, Finland
| | - Antti J Sajanti
- Neurocenter, Department of Neurosurgery, Turku University Hospital and University of Turku, Turku, Finland
| | - Eero A Silver
- Department of Anaesthesia and Intensive Care, University of Turku and Turku University Hospital, Turku, Finland
| | | | - Anton S Bondfolk
- Department of Clinical Medicine, University of Turku, Turku, Finland
| | - Janne Koskimäki
- Neurocenter, Department of Neurosurgery, Turku University Hospital and University of Turku, Turku, Finland
| | - Antti J Saarinen
- Department of Paediatric Orthopaedic Surgery, University of Turku and Turku University Hospital, Kiinamyllynkatu 4-8, 20521, Turku, Finland.
- Department of Orthopaedics and Traumatology, University of Helsinki and Helsinki University Hospital, Helsinki, Finland.
| |
Collapse
|
5
|
Ponkilainen VT, Partio N, Salonen EE, Laine HJ, Mäenpää HM, Mattila VM, Haapasalo HH. Outcomes after nonoperatively treated non-displaced Lisfranc injury: a retrospective case series of 55 patients. Arch Orthop Trauma Surg 2021; 141:1311-1317. [PMID: 32960309 PMCID: PMC8295070 DOI: 10.1007/s00402-020-03599-w] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 02/19/2020] [Accepted: 09/09/2020] [Indexed: 11/25/2022]
Abstract
BACKGROUND Current knowledge of the role of the nonoperative treatment of Lisfranc injuries is based on a few retrospective case series. Hence, consensus on which patients can be treated nonoperatively does not exist. The aim of this study was to investigate outcomes after nonoperative treatment of Lisfranc injuries. METHODS In this study, patients were collected by recruiting all computer tomography-confirmed Lisfranc injuries treated during a 5-year period at a major trauma hospital. Between 2 and 6 years after suffering the injury, patients completed the visual analogue scale foot and ankle questionnaire. RESULTS In total, 55 patients returned adequately completed questionnaires and were included in the study. Of those, 22 patients had avulsion fractures and 33 had simple non-displaced intra-articular fractures. Of these patients, 30 (55%) scored over 90 points in both the pain and function subscales of the VAS-FA, and 35 (64%) scored over 90 points overall. In addition, three (5%) patients scored under 60 points in both the pain and function subscales of the VAS-FA, and four (7%) scored under 60 points overall. Only one patient with avulsion fractures underwent secondary surgery. CONCLUSION Nonoperative treatment has a role in the treatment of Lisfranc injuries, and the results of our study support the view that avulsion and simple intra-articular fractures with < 2 mm of displacement can be treated nonoperatively with high functional outcomes. The results of nonoperative and operative treatment should be compared in a prospective randomized controlled study setting in future studies. LEVEL OF EVIDENCE IV, retrospective case series.
Collapse
Affiliation(s)
- Ville T Ponkilainen
- Department of Orthopaedics, Faculty of Medicine and Life Sciences and Tampere University Hospital, University of Tampere, Teiskontie 35, PL2000, 33521, Tampere, Finland.
- University of Tampere, School of Medicine, 33520, Tampere, Finland.
| | - Nikke Partio
- Department of Orthopaedics, Faculty of Medicine and Life Sciences and Tampere University Hospital, University of Tampere, Teiskontie 35, PL2000, 33521, Tampere, Finland
| | - Essi E Salonen
- Department of Orthopaedics, Faculty of Medicine and Life Sciences and Tampere University Hospital, University of Tampere, Teiskontie 35, PL2000, 33521, Tampere, Finland
| | | | - Heikki M Mäenpää
- Department of Orthopaedics, Faculty of Medicine and Life Sciences and Tampere University Hospital, University of Tampere, Teiskontie 35, PL2000, 33521, Tampere, Finland
| | - Ville M Mattila
- Department of Orthopaedics, Faculty of Medicine and Life Sciences and Tampere University Hospital, University of Tampere, Teiskontie 35, PL2000, 33521, Tampere, Finland
- COXA Hospital for Joint Replacement, Biokatu 6, 33520, Tampere, Finland
| | - Heidi H Haapasalo
- Department of Orthopaedics, Faculty of Medicine and Life Sciences and Tampere University Hospital, University of Tampere, Teiskontie 35, PL2000, 33521, Tampere, Finland
| |
Collapse
|