Machine learning algorithms for predicting scapular kinematics.

The goal of this study was to develop and validate a non-invasive approach to estimate scapular kinematics in individual patients. We hypothesized that machine learning algorithms could be developed using motion capture data to accurately estimate dynamic scapula orientation based on measured humeral orientations and acromion process positions. The accuracy of the algorithms was evaluated against a gold standard of biplane fluoroscopy using a 2D to 3D fluoroscopy/model matching process. Individualized neural networks were developed for nine healthy adult shoulders. These models were used to predict scapulothoracic kinematics, and the predicted kinematics were compared to kinematics obtained using biplane fluoroscopy to determine the accuracy of the machine learning algorithms. Results showed correlations between predicted kinematics and validation kinematics. Estimated kinematics were within 10 of validation kinematics. We concluded that individualized machine learning algorithms show promise for providing accurate, non-invasive measurements of scapulothoracic kinematics.

Validation of a mathematical approach to estimate dynamic scapular orientation.

The goal of this study was to develop and validate a non-invasive approach to estimate scapular kinematics in individual patients. We hypothesized that individualized mathematical algorithms can be developed using motion capture data to accurately estimate dynamic scapula orientation based on measured humeral orientations and acromion process positions. The accuracy of the mathematical algorithms was evaluated against a gold standard of biplane fluoroscopy using a 2D to 3D fluoroscopy/model matching process. Individualized linear models were developed for nine healthy adult shoulders. These models were used to predict scapulothoracic kinematics, and the predicted kinematics were compared to kinematics obtained using biplane fluoroscopy to determine the accuracy of the algorithms. Results showed strong correlations between mathematically predicted kinematics and validation kinematics. Estimated kinematics were within 8° of validation kinematics. We concluded that individualized linear models show promise for providing accurate, non-invasive measurements of scapulothoracic kinematics in a clinical environment.

Reductions in High-End Imaging Utilization With Radiology Review and Consultation.

Following the uptake of value-based purchasing in concert with health care reform in the United States, providers, insurers, and patients are looking for ways to reduce excessive, dangerous, and/or inappropriate high-end imaging utilization (HEIU). Inappropriate HEIU is associated with patient safety risks due to unnecessary exposure to radiation, misappropriation of scarce equipment resources and staff, complications to clinical care, and needless, excessive costs for the patient, hospital, and payer. This paper presents a cost-effective radiology-initiated improvement program piloted in the Christiana Hospital Coordinated Care Network. The pilot demonstrated the effectiveness of regulating high-end imaging orders through radiologists' review of requests of the order as part of the consult process. Over the 2014-2015 fiscal year, 2,177 high-end imaging orders were reviewed by 26 radiologists for approval, rejection, or recommendation of an alternate examination. Of the orders, 86.7% (1887) were approved, 4.0% (87) were rejected, and 9.3% (203) received recommendation for an alternate examination. Based on improved patient safety, cost savings, and appropriate resource use, these findings suggest that radiologists' review can effectively reduce excessive HEIU. This method, with an appropriate algorithm to assist with handling a larger volume of orders, would be ideal to implement systemwide to manage HEIU cost efficiency, simultaneously providing radiologists with more control in their area of expertise and positively impacting quality, safety, and value-based purchasing goals.

Sinus histiocytosis with massive lymphadenopathy (Rosai-Dorfman disease) of the true vocal cord.

Rosai-Dorfman disease (RDD) is a rare disorder involving massive benign proliferation of histiocytes in the lymph nodes. The etiology is unknown and the disease usually presents as painless lymphadenopathy accompanied with fever, night sweats, and weight loss. Due to its rarity, the disease is still difficult to diagnose when associated with extranodal involvement. We present an unusual case of a 39-year-old African American female with RDD involving the left true vocal cord.

Volumetric MRI and MRS and early motor development of infants born preterm.

PURPOSE: To investigate the relationship between volumetric magnetic resonance imaging (MRI) and magnetic resonance spectroscopy (MRS) in infants born less than 30 weeks gestation and early motor development at 6 months adjusted age. METHODS: Twenty infants born preterm and 4 born at term (control) underwent MRI with volumetric analysis and MRS prior to neonatal intensive care unit discharge. Infants were assessed using the Bayley Scale of Infant Development at 6 months adjusted age. RESULTS: At 6 months, infants born preterm with low motor scores had a reduction in their subcortical gray matter. No differences were detected in other brain structures. N-acetylaspartate/choline correlated with white matter (R = 0.45, P = .03), gray matter (R = 0.43, P = .04), and cerebellar volume (R = 0.6, P = .002) but not with 6-month motor performance. CONCLUSION: There is an association between diminished subcortical gray matter volume and low motor scores. Our data suggest that volumetric MRI performed prior to hospital discharge may have some role in counseling parents about potential motor delays.