Kelly, Bryan T.

AbstractDeep learning has a profound impact on daily life. As Orthopedics makes use of this rapid escalation in technology, Orthopedic surgeons will need to take leadership roles on deep learning projects. Moreover, surgeons must possess an understanding of what is necessary to design and implement deep learning‐based project pipelines. This review provides a practical guide for the Orthopedic surgeon to understand the steps needed to design, develop, and deploy a deep learning pipeline for clinical applications. A detailed description of the processes involved in defining the problem, building the team, acquiring and curating the data, labeling the data, establishing the ground truth, pre‐processing and augmenting the data, and selecting the required hardware is provided. In addition, an overview of unique considerations involved in the training and evaluation of deep learning models is provided. This review strives to provide surgeons with the groundwork needed to identify gaps in the clinical landscape that deep learning models may be able to fill and equips them with the knowledge needed to lead an interdisciplinary team through the process of creating novel deep‐learning‐based solutions to fill those gaps.

Abstract Introduction Previous studies have shown that sleep disordered breathing (SDB) is common in patients with Interstitial Lung Disease (ILD), and oximetry is often used for screening prior to further diagnostic testing. Current guidelines recommend polysomnography (PSG) for diagnosis of SDB in patients with significant pulmonary disease, however, home sleep apnea tests (HSAT) are increasingly used in clinical practice for a variety of reasons despite lack of evidence regarding accuracy in this population. In this study, we evaluate the correlation between screening oximetry, a commercial brand HSAT (WatchPAT®) and PSG to examine the diagnostic accuracy of this HSAT technology in patients with ILD. Methods The institution electronic medical record was screened for patients with a diagnosis code for ILD who underwent screening oximetry followed by PSG or HSAT using peripheral arterial tonometry from July 1, 2012 to present. Clinical review confirmed presence of ILD according to American Thoracic Society guidelines. Among the respective cohorts, Paired Wilcoxon Test was used to compare the oximetry 4% oxygen desaturation index (ODI) to the HSAT ODI and PSG apnea-hypopnea index (AHI) as well as percent time spent below oxyhemoglobin saturation of 89%. Spearman correlation was used to correlate the oximetry ODI and parameters of SDB on HSAT and PSG. Results Data was analyzed for 25 patients who had undergone oximetry/HSAT and for 25 patients who had undergone oximetry/PSG. Oximetry ODI showed no significant difference from PSG AHI (p = 0.2635) or between HSAT ODI (p = 0.0755), and no difference was seen in hypoxic time between oximetry and PSG (p = 0.9789). Hypoxic time on HSAT was significantly longer than that on oximetry (p < 0.001). Using HSAT ODI as the standard, HSAT AHI and respiratory disturbance index (RDI) showed rs of 0.9638 and 0.8913 respectively, while oximetry ODI was 0.3893. Compared to PSG AHI, the PSG RDI and oximetry ODI rs were 0.9759 and 0.7407 respectively. Conclusion Among patients with ILD, screening oximetry appears to correlate more strongly with indices of SDB and hypoxic time on PSG rather than HSAT. Further studies are warranted to evaluate efficacy of additional HSAT testing modalities in this patient population. Support (If Any)

See Letter

ABSTRACT Background and objective Criteria for interstitial pneumonia with autoimmune features (IPAF) were recently established for research purposes in a joint statement from the European Respiratory Society (ERS) and American Thoracic Society (ATS). We reviewed the utility of these criteria in patients previously diagnosed as broadly defined undifferentiated connective tissue disease (UCTD) and noted overlapping IPAF findings. Additional review was given to IPAF patients with usual interstitial pneumonia (UIP) on histopathology or radiology in terms of survival and outcome. Methods Patients with prior UCTD‐interstitial lung disease (ILD) were screened by ERS/ATS criteria for IPAF. Clinical data along with all‐cause mortality were collated and compared with selected idiopathic pulmonary fibrosis (IPF) patients from the same study period. Survival was compared between IPAF subgroups with and without UIP features. Results One hundred and one UCTD‐ILD subjects (91%) evaluated from 2005 to 2012 also met strict criteria for IPAF. Frequent clinical findings included Raynaud’s phenomenon, positive anti‐nuclear antibody (ANA) and non‐specific interstitial pneumonia (NSIP) pattern on chest computed tomography (CT). Nineteen had features of UIP either on histopathology or CT imaging. As compared with IPF, IPAF patients had overall better survival except in those with UIP features. Conclusion Current IPAF criteria encompassed the majority of broadly defined UCTD‐ILD and included those with UIP findings. Survival compared with IPF in those with UIP was similar. Further studies are necessary to refine IPAF definitions for clinical use and guide directed management strategies.