Style Transfer-assisted Deep Learning Method for Kidney Segmentation at Multiphase MRI.

Purpose: To develop and validate a semisupervised style transfer-assisted deep learning method for automated segmentation of the kidneys using multiphase contrast-enhanced (MCE) MRI acquisitions. Materials and Methods: This retrospective, Health Insurance Portability and Accountability Act-compliant, institutional review board-approved study included 125 patients (mean age, 57.3 years; 67 male, 58 female) with renal masses. Cohort 1 consisted of 102 coronal T2-weighted MRI acquisitions and 27 MCE MRI acquisitions during the corticomedullary phase. Cohort 2 comprised 92 MCE MRI acquisitions (23 acquisitions during four phases each, including precontrast, corticomedullary, early nephrographic, and nephrographic phases). The kidneys were manually segmented on T2-weighted images. A cycle-consistent generative adversarial network (CycleGAN) was trained to generate anatomically coregistered synthetic corticomedullary style images using T2-weighted images as input. Synthetic images for precontrast, early nephrographic, and nephrographic phases were then generated using the synthetic corticomedullary images as input. Mask region-based convolutional neural networks were trained on the four synthetic phase series for kidney segmentation using T2-weighted masks. Segmentation performance was evaluated in a different cohort of 20 originally acquired MCE MRI examinations by using Dice and Jaccard scores. Results: The CycleGAN network successfully generated anatomically coregistered synthetic MCE MRI-like datasets from T2-weighted acquisitions. The proposed deep learning approach for kidney segmentation achieved high mean Dice scores in all four phases of the original MCE MRI acquisitions (0.91 for precontrast, 0.92 for corticomedullary, 0.91 for early nephrographic, and 0.93 for nephrographic). Conclusion: The proposed deep learning approach achieved high performance in kidney segmentation on different MCE MRI acquisitions.Keywords: Kidney Segmentation, Generative Adversarial Network, CycleGAN, Convolutional Neural Network, Transfer Learning Supplemental material is available for this article. Published under a CC BY 4.0 license.

Predictive Power of Distal Radial Metaphyseal Tenderness for Diagnosing Occult Fracture.

PURPOSE: To correlate the physical examination finding of distal radial metaphyseal tenderness with plain radiographic and magnetic resonance imaging after acute wrist injury to diagnose occult distal radius fractures. We hypothesized that persistent distal radial metaphyseal tenderness 2 weeks after acute injuries is predictive of an occult fracture. METHODS: Twenty-nine adult patients presented, after acute trauma, with distal radial metaphyseal tenderness and initial plain radiographs and/or fluoroscopic images that did not show a distal radius fracture. Patients were reevaluated clinically and radiographically at approximately 2 weeks after initial presentation. Patients with persistent distal radial tenderness and negative radiographs underwent magnetic resonance imaging to definitively diagnose an occult distal radius fracture. We calculated the sensitivity and positive predictive value for persistent distal radial metaphyseal tenderness using a 95% confidence interval and standard formulas. Both radiographs and magnetic resonance imaging were used as our endpoint diagnosis for a distal radius fracture. RESULTS: We diagnosed 28 occult distal radius fractures, 8 by follow-up radiograph and 20 by magnetic resonance imaging. The positive predictive value for patients who completed the protocol was 96%. One patient who did not have an occult distal radius fracture had a fracture of the ulnar styloid. CONCLUSIONS: Tenderness of the distal radial metaphysis after wrist injury is strongly suggestive of a distal radius fracture despite both normal plain radiographs and fluoroscopic images. TYPE OF STUDY/LEVEL OF EVIDENCE: Diagnostic III.

Lack of Acute Hand Care in the Southwest United States.

OBJECTIVES: To describe the epidemiology of acute hand injuries and hand infections and to describe the factors associated with the transfer of these patients to a level 1 trauma center. In addition, we sought to understand management before transfer. DESIGN: Retrospective review of patients with hand trauma or hand infection transferred to our level 1 trauma center from May 2009 to August 2011. We also identified hospitals with emergency departments (EDs) in our region and surveyed ED providers in these hospitals with regard to acute hand care. SETTING: A level 1 trauma center in the United States. PATIENTS: Four hundred sixty consecutive transfers for acute hand care. RESULTS: The average patient age was 38. Most were male (84%), uninsured (51%), and from another county (59%). The average distance of transfer was 51 miles, and 80% were transferred by ground ambulance. The most common reasons for transfer were amputations (24%), infections (21%), lacerations (17%), and fractures/dislocations (16%). Of the 345 hospitals with an ED surveyed, 71% never had hand surgery coverage. CONCLUSIONS: Patients transferred for acute hand care were young and male, and traveled an average 51 miles to get to our center. More than half of these patients were treated and discharged from our ED. This indicates that a majority may have been managed in a clinic setting. Most EDs in our region do not have a hand surgeon available. Most emergency physicians surveyed had received little training in management of acute hand injuries and hand infections. Further research is needed to identify methods to remove barriers to provision of care for patients with hand trauma. LEVEL OF EVIDENCE: Prognostic Level IV. See Instructions for Authors for a complete description of levels of evidence.

Arthroscopically detected intra-articular lesions associated with acute ankle fractures.

BACKGROUND: Anatomic surgical realignment of ankle fractures may still be associated with poor clinical outcomes, possibly as a result of occult intra-articular injury. The aim of this study was to determine if the severity of an acute ankle fracture is correlated with an increased number of arthroscopically detected intra-articular chondral lesions. METHODS: We conducted a retrospective review of the medical charts on 283 ankle fractures that had been treated with open reduction and internal fixation and for which ankle arthroscopy had been routinely performed. The severity of the ankle fractures was categorized, with use of the arthroscopic findings derived from the operative reports as well as the findings on preoperative radiographs, according to the Lauge-Hansen criteria. RESULTS: Of the 283 patients, eighty-four (forty-four female and forty male) met our inclusion criteria. Chondral lesions were found in sixty-one patients (73%). Of seventeen fractures graded as pronation-external rotation or supination-external rotation type I according to the Lauge-Hansen classification, fifteen were associated with one or no chondral lesion and two, with two or more chondral lesions. Of ten fractures graded as pronation-external rotation or supination-external rotation type II, nine were associated with one or no chondral lesion and one, with two or more chondral lesions. Of fifty-six fractures graded as pronation-external rotation or supination-external rotation type IV, twenty-seven were associated with one or no chondral lesion and twenty-nine, with two or more chondral lesions. Type-IV pronation-external rotation and supination-external rotation ankle fractures were more likely to be associated with two or more chondral lesions than type-I fractures (odds ratio = 8.1, 95% confidence interval = 1.7 to 38.6; p = 0.0044) or type-II fractures (odds ratio = 9.7, 95% confidence interval = 1.1 to 81.5; p = 0.0172). CONCLUSIONS: Chondral lesions are commonly found after an acute ankle fracture. This retrospective study demonstrated that the number of intra-articular chondral lesions associated with the more severe ankle fracture patterns (pronation-external rotation and supination-external rotation type-IV fractures) was greater than the number associated with the less severe ankle fracture patterns.