Machine Learning Algorithms to Differentiate Among Pulmonary Complications After Hematopoietic Cell Transplant

Abstract

BACKGROUND: Pulmonary complications, including infections, are highly prevalent in patients after hematopoietic cell transplantation with chronic graft-vs-host disease. These comorbid diseases can make the diagnosis of early lung graft-vs-host disease (bronchiolitis obliterans syndrome) challenging. A quantitative method to differentiate among these pulmonary diseases can address diagnostic challenges and facilitate earlier and more targeted therapy. STUDY DESIGN AND METHODS:We conducted a single-center study of 66 patients with CT chest scans analyzed with a quantitative imaging tool known as parametric response mapping. Parametric response mapping results were correlated with pulmonary function tests and clinical characteristics. Five parametric response mapping metrics were applied to K-means clustering and support vector machine models to distinguish among posttransplantation lung complications solely from quantitative output.RESULTS: Compared with parametric response mapping, spirometry showed a moderate correlation with radiographic air trapping, and total lung capacity and residual volume showed a strong correlation with radiographic lung volumes. K-means clustering analysis distinguished four unique clusters. Clusters 2 and 3 represented obstructive physiology (encompassing 81% of patients with bronchiolitis obliterans syndrome) in increasing severity (percentage air trapping 15.6% and 43.0%, respectively). Cluster 1 was dominated by normal lung, and cluster 4 was characterized by patients with parenchymal opacities. A support vector machine algorithm differentiated bronchiolitis obliterans syndrome with a specificity of 88%, sensitivity of 83%, accuracy of 86%, and an area under the receiver operating characteristic curve of 0.85. INTERPRETATION:Our machine learning models offer a quantitative approach for the identification of bronchiolitis obliterans syndrome vs other lung diseases, including late pulmonary complications after hematopoietic cell transplantation.

A Systematic Review of Machine Learning Techniques in Hematopoietic Stem Cell Transplantation (HSCT)

“…Lopez et al [29] used random forest to identify genes associated with the incidence of chronic GVHD while Sharifi et al [30] used unsupervised methods to distinguish pulmonary complications post-HSCT. Gandelman et al [31] also utilized unsupervised machine learning techniques for risk stratification of chronic GVHD.…”

Quantitative Lung Imaging in Hematopoietic Cell Transplant Recipients

“…Their identification of numerous HCT recipients with BOS, all of whom underwent standardized CT image acquisition and analysis, is an arduous and painstaking task. Sharifi et al 11 establish the feasibility of advanced image analysis with PRM and similar quantitative techniques to detect clinically evident BOS. The ambition of investigators should now be to test the utility of quantitative CT imaging in patients with early pulmonary impairment after HCT, which will require more intensive pulmonary screening of HCT recipients than currently performed.…”

“…In this issue of CHEST, Sharifi et al 11 examine the utility of PRM to evaluate the cause of dyspnea in 66 HCT recipients. Standardized protocols were used to obtain inspiratory and expiratory CT scans, and PRM was applied to these images to quantify the degree of gas trapping.…”

“…The current study is just the first step in achieving that goal. What Sharifi et al 11 have shown us is that a quantitative image analysis can measure many of the same parameters as pulmonary function testing, while simultaneously evaluating for infection, organizing pneumonia, or truncal sclerosis; it is, essentially, a "one-stop shop" to evaluate for key pulmonary complications after HCT. The next target will be to test these quantitative markers in early pulmonary impairment, where prompt intervention may be more likely to result in stabilization of pulmonary function.…”

Head-to-head Comparison of Qualitative Radiologist Assessment With Automated Quantitative Computed Tomography Analysis for Bronchiolitis Obliterans Syndrome After Hematopoietic Cell Transplantation

“…8 These were compared to the gold-standard clinician diagnosis by NIH criteria. 13 A subset of these data with differing diagnostic adjudication and disease classification was used in a previous machine learning study by our group 22 (see Supplemental Data File for details, Supplemental Digital Content 1, http://links.lww.com/JTI/ A190). The association of pertinent imaging findings was also analyzed with respect to PRM metrics, thus offering an integrated approach for quantitative and qualitative assessments to enhance clinician and radiologist diagnosis of BOS.…”