RÉSUMÉ
PURPOSE: To develop a deep learning (DL) model that can simultaneously detect lateral and medial collateral ligament injuries of the ankle, aiding in the diagnosis of chronic ankle instability (CAI), and assess its impact on clinicians' diagnostic performance. METHODS: DL models were developed and external validated on retrospectively collected ankle MRIs between April 2016 and March 2022 respectively at three centers. Included patients were confirmed diagnoses of CAI through arthroscopy, as well as individuals who had undergone MRI and physical examinations that ruled out ligament injuries. DL models were constructed based on a multi-label paradigm. A transformer-based multi-label DL model (AnkleNet) was developed and compared with four convolution neural network (CNN) models. Subsequently, a reader study was conducted to evaluate the impact of model assistance on clinicians when diagnosing challenging cases: identifying rotational CAI (RCAI). Diagnostic performance was assessed using area under the receiver operating characteristic curve (AUC). RESULTS: Our transformer-based model achieved AUC of 0.910 and 0.892 for detecting lateral and medial collateral ligament injury, respectively, both of which was significantly higher than that of CNN-based models (all P < 0.001). In terms of further CAI diagnosis, it exhibited a macro-average AUC of 0.870 and a balanced accuracy of 0.805. The reader study indicated that incorporation with our model significantly enhanced the diagnostic accuracy of clinicians (P = 0.042), particularly junior clinicians, and led to a reduction in diagnostic variability. The code of the model can be accessed at https://github.com/ChiariRay/AnkleNet. CONCLUSION: Our transformer-based model was able to detect lateral and medial collateral ligament injuries based on MRI and outperformed CNN-based models, demonstrating a promising performance in diagnosing CAI, especially RCAI patients.
RÉSUMÉ
OBJECTIVE: To investigate the clinical outcomes of the use of the perforator propeller flaps based on the end dorsal branch of digital proper artery in the same finger in repair of finger pulp defects and sensory reconstruction in children. METHODS: Twenty-three children (31 fingers) with index, middle, ring or little finger pulp defects were hospitalized from September 2012 to December 2013. The area of finger pulp defects ranged from 1.2 cm × 1.0 cm to 2.0 cm × 1.5 cm. The perforator propeller flaps based on the end dorsal branch of digital proper artery in the same finger were used to repair the defects, with the flap size ranging from 1.3 cm × 1.2 cm to 2.2 cm × 1.6 cm. The dorsal branch of the digital proper nerve of the flap was conducted end-to-end anastomosis with the broken end of the nerve of the wound to reconstruct sensation. The donor sites were covered with autologous full-thickness skin obtained from inner aspect of the thigh. RESULTS: Primary healing of the wounds and donor sites were achieved in all 23 children. All the flaps and skin grafts of donor sites survived. All the patients were followed up for 6 to 20 months, with mean time of 14 months. At the last follow-up, the flaps and donor sites were in good appearance, the finger pulps were mellow and plump, with no obvious pigmentation or cicatricial contracture. The sensation of finger pulps reached S3(+), and the distance of two-point discrimination ranged from 4.5 to 6.0 mm, with mean distance of 5.1 mm. Twenty-one parents of the patients were strongly satisfied with the appearance of the repaired fingers, and the other 2 parents also expressed satisfaction. CONCLUSIONS: Transplantation of the perforator propeller flap based on the end dorsal branch of digital proper artery in the same finger is a safe and reliable method for the repair of index, middle, ring, and little finger pulp defects and sensory reconstruction of children. The flaps are with good blood supply, appearance and sensory function after operation.
