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

Abstract

Purpose: Computed tomography (CT) findings of bronchiolitis obliterans syndrome (BOS) can be nonspecific and variable. This study aims to measure the incremental value of automated quantitative lung CT analysis to clinical CT interpretation. A head-to-head comparison of quantitative CT lung density analysis by parametric response mapping (PRM) with qualitative radiologist performance in BOS diagnosis was performed.Materials and Methods: Inspiratory and end-expiratory CTs of 65 patients referred to a post-bone marrow transplant lung graftversus-host-disease clinic were reviewed by 3 thoracic radiologists for the presence of mosaic attenuation, centrilobular opacities, airways dilation, and bronchial wall thickening. Radiologists' majority consensus diagnosis of BOS was compared with automated PRM air trapping quantification and to the gold-standard diagnosis of BOS as per National Institutes of Health (NIH) consensus criteria.Results: Using a previously established threshold of 28% air trapping on PRM, the diagnostic performance for BOS was as follows: sensitivity 56% and specificity 94% (area under the receiver operator curve [AUC] = 0.75). Radiologist review of inspiratory CT images alone resulted in a sensitivity of 80% and a specificity of 69% (AUC = 0.74). When radiologists assessed both inspiratory and end-expiratory CT images in combination, the sensitivity was 92% and the specificity was 59% (AUC = 0.75). The highest performance was observed when the quantitative PRM report was reviewed alongside inspiratory and end-expiratory CT images, with a sensitivity of 92% and a specificity of 73% (AUC = 0.83). Conclusions:In the CT diagnosis of BOS, qualitative expert radiologist interpretation was noninferior to quantitative PRM. The highest level of diagnostic performance was achieved by the combination of quantitative PRM measurements with qualitative image feature assessments.

Harnessing Deep Learning to Detect Bronchiolitis Obliterans Syndrome from Chest CT

“…Previous methods of CT scan analysis for BOS detection rely on pre-defined procedures to identify patterns in the lung and cannot benefit from signs of BOS not captured by these procedures. 10,11,12,13,14,15,16,17,18 However, designing procedures to extract all pertinent information from the scans may not be feasible. Prior research suggests that BOS can manifest itself in non-obvious ways, including by alterations to the volume and surface of the airways, modifications of the volume of pulmonary vasculature 24,17 , and changes in the density of microvasculature 25,26 , all of which may affect the appearance of the scans.…”

“…Ongoing research on computational methods to detect BOS from CT scans were designed to address the current limitations of CT imaging in BOS diagnosis. However, previous methods 10,11,12,13,14,15,16,17,18 rely on heavily engineered procedures for extracting information from the scans, and designing these procedures to maximize robustness in scans taken at different resolutions or respiration phases is not feasible. These drawbacks can be overcome by using a deep neural network (DNN) 19 , which have thus far not been used for BOS detection.…”

Imaging of pediatric hematopoietic stem cell transplant recipients: A COG Diagnostic Imaging Committee/SPR Oncology Committee White Paper

“…Chest CT is recommended for imaging evaluation of pulmonary GVHD and drug-induced pulmonary toxicities (Grade B; SOR 1.56, strong recommendation). 39,41,42,44,[66][67][68][69][70] On chest CT, GVHD can result in two distinct interstitial lung disease patterns: bronchiolitis obliterans and organizing pneumonia (Table 6).…”