Jonathan G. Goldin’s research while affiliated with University of California, Los Angeles and other places

Background Deep learning (DL)‐based systems have not yet been broadly implemented in clinical practice, in part due to unknown robustness across multiple imaging protocols. Purpose To this end, we aim to evaluate the performance of several previously developed DL‐based models, which were trained to distinguish idiopathic pulmonary fibrosis (IPF) from non‐IPF among interstitial lung disease (ILD) patients, under standardized reference CT imaging protocols. In this study, we utilized CT scans from non‐IPF ILD subjects, acquired using various imaging protocols, to assess the model performance. Methods Three DL‐based models, including one 2D and two 3D models, have been previously developed to classify ILD patients into IPF or non‐IPF based on chest CT scans. These models were trained on CT image data from 389 IPF and 700 non‐IPF ILD patients, retrospectively, obtained from five multicenter studies. For some patients, multiple CT scans were acquired (e.g., one at inhalation and one at exhalation) and/or reconstructed (e.g., thin slice and/or thick slice). Thus, for each patient, one CT image dataset was selected to be used in the construction of the classification model, so the parameters of that data set serve as the reference conditions. In one non‐IPF ILD study, due to its specific study protocol, many patients had multiple CT image data sets that were acquired under both prone and supine positions and/or reconstructed under different imaging parameters. Therefore, to assess the robustness of the previously developed models under different (e.g., non‐reference) imaging protocols, we identified 343 subjects from this study who had CT data from both the reference condition (used in model construction) and non‐reference conditions (e.g., evaluation conditions), which we used in this model evaluation analysis. We reported the specificities from three model under the non‐reference conditions. Generalized linear mixed effects model (GLMM) was utilized to identify the significant CT technical and clinical parameters that were associated with getting inconsistent diagnostic results between reference and evaluation conditions. Selected parameters include effective tube current‐time product (known as “effective mAs”), reconstruction kernels, slice thickness, patient orientation (prone or supine), CT scanner model, and clinical diagnosis. Limitations include the retrospective nature of this study. Results For all three DL models, the overall specificity of the previously trained IPF diagnosis model decreased (p < 0.05 for two out of three models). GLMM further suggests that for at least one out of three models, mean effective mAs across the scan is the key factor that leads to the decrease in model predictive performance (p < 0.001); the difference of mean effective mAs between the reference and evaluation conditions (p = 0.03) and slice thickness (3 mm; p = 0.03) are flagged as significant factors for one out of three models; other factors are not statistically significant (p > 0.05). Conclusion Preliminary findings demonstrated the lack of robustness of IPF diagnosis model when the DL‐based model is applied to CT series collected under different imaging protocols, which indicated that care should be taken as to the acquisition and reconstruction conditions used when developing and deploying DL models into clinical practice.

High-resolution CT (HRCT) plays an important role in diagnosing and monitoring interstitial lung diseases (ILDs). Despite advances, predicting disease progression and treatment response remains challenging. HRCT enables noninvasive visualization and classification of patterns of lung injury and assessment of disease extent. Visual estimation of CT extent of fibrotic lung disease is an independent predictor of mortality and progression, but is subjective, with only modest interobserver agreement for radiologic interpretation of ILD. Machine learning-based textural analysis of fibrosis extent on baseline and serial HRCT scans shows robust correlations with physiologic measures and strong association with risk of disease progression or mortality across various fibrosing ILDs. In idiopathic pulmonary fibrosis, quantitative CT (QCT) assessment is associated with physiologic impairment and risk of progression and death, and increasing severity of fibrosis on longitudinal evaluation is associated with increased risk of progression and death. Similar results have been noted for fibrotic hypersensitivity pneumonitis and connective tissue disease. This review focuses on QCT techniques for ILDs. We describe the clinical need for quantification of lung disease and illustrate the role of conventional visual evaluation and of QCT approaches in defining disease severity, prognosis, and longitudinal progression, both in established disease and in preclinical interstitial abnormality.

The Response Evaluation in Solid Tumors (RECIST) 1.1 provides key guidance for performing imaging response assessment and defines image-based outcome metrics in oncology clinical trials, including progression free survival. In this framework, tumors identified on imaging are designated as either target lesions, non-target disease or new lesions and a structured categorical response is assigned at each imaging time point. While RECIST provides definitions for these categories, it specifically and objectively defines only the target disease. Predefined thresholds of size change provide unbiased metrics for determining objective response and disease progression of the target lesions. However, worsening of non-target disease or emergence of new lesions is given the same importance in determining disease progression despite these being qualitatively assessed and less rigorously defined. The subjective assessment of non-target and new disease contributes to reader variability, which can impact the quality of image interpretation and even the determination of progression free survival. The RECIST Working Group has made significant efforts in developing RECIST 1.1 beyond its initial publication, particularly in its application to targeted agents and immunotherapy. A review of the literature highlights that the Working Group has occasionally employed or adopted objective measures for assessing non-target and new lesions in their evaluation of RECIST-based outcome measures. Perhaps a prospective evaluation of these more objective definitions for non-target and new lesions within the framework of RECIST 1.1 might improve reader interpretation. Ideally, these changes could also better align with clinically meaningful outcome measures of patient survival or quality of life.

Background Systemic sclerosis (SSc) is a complex autoimmune disease characterized by inflammation and fibrosis across different organ systems, including the gastrointestinal tract and the lungs. Studies have demonstrated that gut microbiota modulate pulmonary immune function [1], and alterations of intestinal microbiota communities can influence disease outcomes in distant organs, including the lungs, through gut transfer experiments of dysbiotic microbiota [2]. To our knowledge, no prior studies have investigated the gut-lung axis in SSc-associated interstitial lung disease (ILD) using an international, multicentre study. Objectives We aimed to identify differentially abundant bacterial species in SSc patients with ILD compared to without ILD and to determine whether specific bacterial species are associated with ILD severity based on the quantitative radiological extent of ILD. Methods SSc patients with and without ILD were recruited from 7 international SSc Centres (University of California, Los Angeles [UCLA], USA; Lund University [LU], Sweden; Duke-National University of Singapore; Johns Hopkins University, USA; Ghent University, Belgium; University of Adelaide, Australia; Pontificia Universidad Católica de Chile) and provided a stool sample. Shotgun metagenomics were performed using the Illumina NovaSeq 6000 with a target depth of 10 million 150x2 sequences per sample. Shotgun reads were inputted into MetaPhlAn4 for taxonomic identification of species for compositional analysis and subsequently underwent center log-ratio transformation. Samples were filtered to retain species with at least 10% non-zero counts. High-resolution computed tomography (HRCT) scans of the chest underwent quantitative image analysis to determine the radiological extent of ILD (QILD) in patients from UCLA and LU. General linear models were applied to identify differentially abundant species based on ILD presence and determine associations between QILD scores and species abundance, adjusting for body mass index, current proton pump inhibitor use, current probiotic use, current or prior immunomodulatory therapy, presence of small intestinal bacterial overgrowth and site. We considered p<0.05 at the threshold for reporting and provide 5% false discovery rate corrected p-values (q). We also computed effect size estimates, Cohen’s D for mean differences and standardized Beta for association analyses. Results Among the 261 SSc participants, 220 (84%) were female and 167 (64%) had HRCT-defined ILD. The mean age was 54.6 (SD 13) years; the mean body mass index (BMI) was 24.9 (SD 4.8); and the median disease duration was 6.8 (IQR 3.5, 12.9) years. Among 254 species analyzed, the abundance of 12 bacterial species was altered in patients with ILD compared to without ILD in all study participants, including Anaerotignum faecicola (Cohen’s d= 0.37 [95% CI 0.07-0.68]; p=0.017; q=0.868) and Roseburia hominis (Cohen’s d= 0.42 [95% CI 0.11-0.72]; p=0.007; q=0.868). Among 113 SSc-ILD participants who had an HRCT scan amenable to quantitative image analysis, 16 bacterial species were associated with severity of ILD based on the QILD scores, including Dysosmobacter welbionis (Standardized beta 0.33; p<0.0001; q=0.34 [Figure 1]) and Bifidobacterium adolescentis (Standardized beta -0.24; p=0.02; q=0.56 [Figure 1]). Conclusion Patients with SSc-ILD recruited from several international centres have a unique microbiota signature compared with SSc patients without ILD. The finding that the abundance of certain bacterial species is associated with ILD and its severity supports the hypothesis that intestinal dysbiosis may contribute to ILD pathogenesis. Future studies are needed to identify additional mediators of the gut-lung axis in SSc-ILD, including bacterial metabolites. REFERENCES [1] Sencio V, et al. Mucosal Immunol 2021;14:296–304.[2] Skalaski JH, et al. PLoS Pathog 2018;14:e1007260. • Download figure • Open in new tab • Download powerpoint Figure 1. Association between the abundance of Dysosmobacter welbionis (top) and Bifidobacterium adolescentis (bottom) and QILD score in SSc-ILD patients from LU in Sweden (red) and UCLA in USA (blue) based on generalized linear model estimates. Acknowledgements Funding Sources: Anonymous donation (EV), NHLBI (EV), Boehringer Ingelheim (EV and KA) Disclosure of Interests Kristofer Andréasson Johnson & Johnson Innovative Medicine, Swapna Joshi: None declared, Jen Labus: None declared, Arissa Young: None declared, Andrea Low: None declared, Vanessa Smith Boehringer Ingelheim, Zsuzsanna McMahan Boehringer Ingelheim, Susanna M. Proudman Boehringer Ingelheim, Janssen, Boehringer Ingelheim, Janssen, Antonia Valenzuela Vegara: None declared, Grace Kim: None declared, Jonathan Goldin: None declared, Ezinne Aja: None declared, Jonathan Jacobs: None declared, Elizabeth Volkmann Boehringer Ingelheim, GSK, Horizon, Prometheus, Boehringer Ingelheim.

Purpose: Evaluation of lung fissure integrity is required to determine whether emphysema patients have complete fissures and are candidates for endobronchial valve (EBV) therapy. We propose a deep learning (DL) approach to segment fissures using a three-dimensional patch-based convolutional neural network (CNN) and quantitatively assess fissure integrity on CT to evaluate it in subjects with severe emphysema. Approach: From an anonymized image database of patients with severe emphysema, 129 CT scans were used. Lung lobe segmentations were performed to identify lobar regions, and the boundaries among these regions were used to construct approximate interlobar regions of interest (ROIs). The interlobar ROIs were annotated by expert image analysts to identify voxels where the fissure was present and create a reference ROI that excluded non-fissure voxels (where the fissure is incomplete). A CNN configured by nnU-Net was trained using 86 CT scans and their corresponding reference ROIs to segment the ROIs of left oblique fissure (LOF), right oblique fissure (ROF), and right horizontal fissure (RHF). For an independent test set of 43 cases, fissure integrity was quantified by mapping the segmented fissure ROI along the interlobar ROI. A fissure integrity score (FIS) was then calculated as the percentage of labeled fissure voxels divided by total voxels in the interlobar ROI. Predicted FIS (p-FIS) was quantified from the CNN output, and statistical analyses were performed comparing p-FIS and reference FIS (r-FIS). Results: The absolute percent error mean (±SD) between r-FIS and p-FIS for the test set was 4.0% (±4.1%), 6.0% (±9.3%), and 12.2% (±12.5%) for the LOF, ROF, and RHF, respectively. Conclusions: A DL approach was developed to segment lung fissures on CT images and accurately quantify FIS. It has potential to assist in the identification of emphysema patients who would benefit from EBV treatment.