Enchondroma Detection from Hand Radiographs with an Interactive Deep Learning Segmentation Tool-A Feasibility Study

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.

DupuytrEn Treatment EffeCtiveness Trial (DETECT): a protocol for prospective, randomised, controlled, outcome assessor-blinded, three-armed parallel 1:1:1, multicentre trial comparing the effectiveness and cost of collagenase clostridium histolyticum, percutaneous needle fasciotomy and limited fasciectomy as short-term and long-term treatment strategies in Dupuytren's contracture

INTRODUCTION

Dupuytren's contracture (DC) is a chronic fibroproliferative disorder of the palmar fascia which leads to flexion contracture in one or more fingers. There is no definitive cure for DC, and treatment aims at relieving symptoms by releasing the contracture using percutaneous or operative techniques.

METHODS AND ANALYSIS

We planned a prospective, randomised, controlled, outcome assessor-blinded, three-armed parallel 1:1:1, multicentre trial comparing the effectiveness and cost of (1) collagenase clostridium histolyticum injection followed by limited fasciectomy in non-responsive cases, (2) percutaneous needle fasciotomy followed by limited fasciectomy in non-responsive cases and (3) primary limited fasciectomy during short-term and long-term follow-up for Tubiana I-III stages DC. We will recruit participants from seven national centres in Finland. Primary outcome is the rate of success in the treatment arm at 5 years after recruitment. Success is a composite outcome comprising (1) at least 50% contracture release from the date of recruitment and (2) participants in a patient-accepted symptom state (PASS). Secondary outcomes are (1) angle of contracture, (2) quick disabilities of the arm, a shoulder and hand outcome measure (QuickDASH), (3) perceived hand function, (4) EQ-5D-3L, (5) rate of major adverse events, (6) patient's trust of the treatment, (7) global rating, (8) rate of PASS, (9) rate of minimal clinically important improvement, (10) expenses, (11) progression of disease, (12) progression-free survival, (13) favoured treatment modality, (14) patients achieving full contracture release and >50% improvement and (15) patient satisfaction with the treatment effect. Predictive factors for achieving the PASS will also be analysed.

ETHICS AND DISSEMINATION

The protocol was approved by the Tampere University Hospital Institutional Review Board and Finnish Medicine Agency. The study will be performed according to the principles of good clinical practice. The results of the trial will be disseminated as published articles in peer-reviewed journals.

TRIAL REGISTRATION NUMBER

NCT03192020; Pre-results.