Quantitative Computed Tomography Lung COVID Scores with Laboratory Markers: Utilization to Predict Rapid Progression and Monitor Longitudinal Changes in Patients with Coronavirus 2019 (COVID-19) Pneumonia

Coronavirus disease 2019 (COVID-19), is an ongoing issue in certain populations, presenting rapidly worsening pneumonia and persistent symptoms. This study aimed to test the predictability of rapid progression using radiographic scores and laboratory markers and present longitudinal changes. This retrospective study included 218 COVID-19 pneumonia patients admitted at the Chungnam National University Hospital. Rapid progression was defined as respiratory failure requiring mechanical ventilation within one week of hospitalization. Quantitative COVID (QCOVID) scores were derived from high-resolution computed tomography (CT) analyses: (1) ground glass opacity (QGGO), (2) mixed diseases (QMD), and (3) consolidation (QCON), and the sum, quantitative total lung diseases (QTLD). Laboratory data, including inflammatory markers, were obtained from electronic medical records. Rapid progression was observed in 9.6% of patients. All QCOVID scores predicted rapid progression, with QMD showing the best predictability (AUC = 0.813). In multivariate analyses, the QMD score and interleukin(IL)-6 level were important predictors for rapid progression (AUC = 0.864). With >2 months follow-up CT, remained lung lesions were observed in 21 subjects, even after several weeks of negative reverse transcription polymerase chain reaction test. AI-driven quantitative CT scores in conjugation with laboratory markers can be useful in predicting the rapid progression and monitoring of COVID-19.

Radiologist and Radiology Practice Wellbeing: A Report of the 2023 ARRS Wellness Summit

In April 2023, the first American Roentgen Ray Society (ARRS) Wellness Summit was held in Honolulu, Hawaii. The Summit was a communal call to action bringing together professionals from the field of radiology to critically review our current state of wellness and reimagine the role of radiology and radiologists to further wellbeing. The in-person and virtual Summit was available free-of-cost to all meeting registrants and included 12 sessions with 44 invited moderators and panelists. The Summit aimed to move beyond simply rehashing the repeated issues and offering theoretical solutions, and instead focus on intentional practice evolution, identifying implementable strategies so that we as a field can start to walk our wellness talk. Here, we first summarize the thematic discussions from the 2023 ARRS Wellness Summit, and second, share several strategic action items that emerged.

Quantitative interstitial lung disease scores in idiopathic inflammatory myopathies: longitudinal changes and clinical implications

OBJECTIVES

To investigate computer-aided quantitative scores from high-resolution CT (HRCT) images and determine their longitudinal changes and clinical significance in patients with idiopathic inflammatory myopathies (IIMs)-related interstitial lung disease (IIMs-ILD).

METHODS

The clinical data and HRCT images of 80 patients with IIMs who underwent serial HRCT scans at least twice were retrospectively analysed. Quantitative ILD (QILD) scores (%) were calculated as the sum of the extent of lung fibrosis, ground-glass opacity, and honeycombing. The individual time-estimated ΔQILD between two consecutive scans was derived using a linear approximation of yearly changes.

RESULTS

The baseline median QILD (interquartile range) scores in the whole lung were 28.1% (19.1-43.8). The QILD was significantly correlated with forced vital capacity (r = -0.349, P = 0.002) and diffusing capacity for carbon monoxide (r = -0.381, P = 0.001). For ΔQILD between the first two scans, according to the visual ILD subtype, QILD aggravation was more frequent in patients with usual interstitial pneumonia (UIP) than non-UIP (80.0% vs 44.4%, P = 0.013). Multivariable logistic regression analyses identified UIP was significantly related to radiographic ILD progression (ΔQILD >2%, P = 0.015). Patients with higher baseline QILD scores (>28.1%) had a higher risk of lung transplantation or death (P = 0.015). In the analysis of three serial HRCT scans (n = 41), dynamic ΔQILD with four distinct patterns (improving, worsening, convex and concave) was observed.

CONCLUSION

QILD changes in IIMs-ILD were dynamic, and baseline UIP patterns seemed to be related to a longitudinal progression in QILD. These may be potential imaging biomarkers for lung function, changes in ILD severity and prognosis in IIMs-ILD.

Lung Cancer Screening Among Mammography Patients: Knowledge, Eligibility, Participation, and Interest

OBJECTIVE

To determine lung cancer screening eligibility, knowledge, and interest and to quantify the effect of the expanded 2021 lung cancer screening eligibility criteria among women presenting for screening mammography, a group with demonstrable interest in cancer screening.

METHODS

A single-page survey was distributed to patients presenting for screening mammography, from January-March 2020 and June 2020-January 2021, at 2 academic medical centers on the East and West Coasts. The population served by the East Coast institution has greater poverty, greater ethnic/racial diversity, and lower education levels. Survey questions included age, smoking history, lung cancer screening knowledge, participation, and interest. Lung cancer screening eligibility was determined for both 2013 and 2021 USPSTF guidelines. Descriptive statistics were calculated, and data were compared between groups using the Chi-square test, Mann-Whitney nonparametric test, and the 2-sample t test.

RESULTS

5512 surveys were completed; 33% (1824) of women reported a history of smoking-30% (1656) former smokers and 3% (156) current smokers. Among women with a smoking history, 7% (127/1824) were eligible for lung cancer screening using 2013% and 11% (207/1824) using the 2021 USPSTF criteria. Interest in lung cancer screening was high (73%; 151/207) among eligible women using 2021 USPSTF criteria, but only 42% (87/207) had heard of lung cancer screening and only 28% (57/207) had received prior LDCT screening.

CONCLUSION

Eligible screening mammography patients reported high levels of interest in lung cancer screening but low levels of knowledge and participation. Linking mammography and LDCT appointments may improve lung cancer screening participation.

Translating AI to Clinical Practice: Overcoming Data Shift with Explainability

To translate artificial intelligence (AI) algorithms into clinical practice requires generalizability of models to real-world data. One of the main obstacles to generalizability is data shift, a data distribution mismatch between model training and real environments. Explainable AI techniques offer tools to detect and mitigate the data shift problem and develop reliable AI for clinical practice. Most medical AI is trained with datasets gathered from limited environments, such as restricted disease populations and center-dependent acquisition conditions. The data shift that commonly exists in the limited training set often causes a significant performance decrease in the deployment environment. To develop a medical application, it is important to detect potential data shift and its impact on clinical translation. During AI training stages, from premodel analysis to in-model and post hoc explanations, explainability can play a key role in detecting model susceptibility to data shift, which is otherwise hidden because the test data have the same biased distribution as the training data. Performance-based model assessments cannot effectively distinguish the model overfitting to training data bias without enriched test sets from external environments. In the absence of such external data, explainability techniques can aid in translating AI to clinical practice as a tool to detect and mitigate potential failures due to data shift. ©RSNA, 2023 Quiz questions for this article are available in the supplemental material.

Automated Endotracheal Tube Placement Check Using Semantically Embedded Deep Neural Networks

RATIONALE AND OBJECTIVES

To develop artificial intelligence (AI) system that assists in checking endotracheal tube (ETT) placement on chest X-rays (CXRs) and evaluate whether it can move into clinical validation as a quality improvement tool.

MATERIALS AND METHODS

A retrospective data set including 2000 de-identified images from intensive care unit patients was split into 1488 for training and 512 for testing. AI was developed to automatically identify the ETT, trachea, and carina using semantically embedded neural networks that combine a declarative knowledge base with deep neural networks. To check the ETT tip placement, a "safe zone" was computed as the region inside the trachea and 3-7 cm above the carina. Two AI outputs were evaluated: (1) ETT overlay, (2) ETT misplacement alert messages. Clinically relevant performance metrics were compared against prespecified thresholds of >85% overlay accuracy and positive predictive value (PPV) > 30% and negative predictive value NPV > 95% for alerts to move into clinical validation.

RESULTS

An ETT was present in 285 of 512 test cases. The AI detected 95% (271/285) of ETTs, 233 (86%) of these with accurate tip localization. The system (correctly) did not generate an ETT overlay in 221/227 CXRs where the tube was absent for an overall overlay accuracy of 89% (454/512). The alert messages indicating that either the ETT was misplaced or not detected had a PPV of 83% (265/320) and NPV of 98% (188/192).

CONCLUSION

The chest X-ray AI met prespecified performance thresholds to move into clinical validation.

Multi-scale, domain knowledge-guided attention + random forest: a two-stage deep learning-based multi-scale guided attention models to diagnose idiopathic pulmonary fibrosis from computed tomography images

BACKGROUND

Idiopathic pulmonary fibrosis (IPF) is a progressive, irreversible, and usually fatal lung disease of unknown reasons, generally affecting the elderly population. Early diagnosis of IPF is crucial for triaging patients' treatment planning into anti-fibrotic treatment or treatments for other causes of pulmonary fibrosis. However, current IPF diagnosis workflow is complicated and time-consuming, which involves collaborative efforts from radiologists, pathologists, and clinicians and it is largely subject to inter-observer variability.

PURPOSE

The purpose of this work is to develop a deep learning-based automated system that can diagnose subjects with IPF among subjects with interstitial lung disease (ILD) using an axial chest computed tomography (CT) scan. This work can potentially enable timely diagnosis decisions and reduce inter-observer variability.

METHODS

Our dataset contains CT scans from 349 IPF patients and 529 non-IPF ILD patients. We used 80% of the dataset for training and validation purposes and 20% as the holdout test set. We proposed a two-stage model: at stage one, we built a multi-scale, domain knowledge-guided attention model (MSGA) that encouraged the model to focus on specific areas of interest to enhance model explainability, including both high- and medium-resolution attentions; at stage two, we collected the output from MSGA and constructed a random forest (RF) classifier for patient-level diagnosis, to further boost model accuracy. RF classifier is utilized as a final decision stage since it is interpretable, computationally fast, and can handle correlated variables. Model utility was examined by (1) accuracy, represented by the area under the receiver operating characteristic curve (AUC) with standard deviation (SD), and (2) explainability, illustrated by the visual examination of the estimated attention maps which showed the important areas for model diagnostics.

RESULTS

During the training and validation stage, we observe that when we provide no guidance from domain knowledge, the IPF diagnosis model reaches acceptable performance (AUC±SD = 0.93±0.07), but lacks explainability; when including only guided high- or medium-resolution attention, the learned attention maps are not satisfactory; when including both high- and medium-resolution attention, under certain hyperparameter settings, the model reaches the highest AUC among all experiments (AUC±SD = 0.99±0.01) and the estimated attention maps concentrate on the regions of interests for this task. Three best-performing hyperparameter selections according to MSGA were applied to the holdout test set and reached comparable model performance to that of the validation set.

CONCLUSIONS

Our results suggest that, for a task with only scan-level labels available, MSGA+RF can utilize the population-level domain knowledge to guide the training of the network, which increases both model accuracy and explainability.

Quantitative Imaging Metrics for the Assessment of Pulmonary Pathophysiology: An Official American Thoracic Society and Fleischner Society Joint Workshop Report

Multiple thoracic imaging modalities have been developed to link structure to function in the diagnosis and monitoring of lung disease. Volumetric computed tomography (CT) renders three-dimensional maps of lung structures and may be combined with positron emission tomography (PET) to obtain dynamic physiological data. Magnetic resonance imaging (MRI) using ultrashort-echo time (UTE) sequences has improved signal detection from lung parenchyma; contrast agents are used to deduce airway function, ventilation-perfusion-diffusion, and mechanics. Proton MRI can measure regional ventilation-perfusion ratio. Quantitative imaging (QI)-derived endpoints have been developed to identify structure-function phenotypes, including air-blood-tissue volume partition, bronchovascular remodeling, emphysema, fibrosis, and textural patterns indicating architectural alteration. Coregistered landmarks on paired images obtained at different lung volumes are used to infer airway caliber, air trapping, gas and blood transport, compliance, and deformation. This document summarizes fundamental "good practice" stereological principles in QI study design and analysis; evaluates technical capabilities and limitations of common imaging modalities; and assesses major QI endpoints regarding underlying assumptions and limitations, ability to detect and stratify heterogeneous, overlapping pathophysiology, and monitor disease progression and therapeutic response, correlated with and complementary to, functional indices. The goal is to promote unbiased quantification and interpretation of in vivo imaging data, compare metrics obtained using different QI modalities to ensure accurate and reproducible metric derivation, and avoid misrepresentation of inferred physiological processes. The role of imaging-based computational modeling in advancing these goals is emphasized. Fundamental principles outlined herein are critical for all forms of QI irrespective of acquisition modality or disease entity.

Detection and Early Referral of Patients With Interstitial Lung Abnormalities: An Expert Survey Initiative

BACKGROUND

Interstitial lung abnormalities (ILA) may represent undiagnosed early-stage or subclinical interstitial lung disease (ILD). ILA are often observed incidentally in patients who subsequently develop clinically overt ILD. There is limited information on consensus definitions for, and the appropriate evaluation of, ILA. Early recognition of patients with ILD remains challenging, yet critically important. Expert consensus could inform early recognition and referral.

RESEARCH QUESTION

Can consensus-based expert recommendations be identified to guide clinicians in the recognition, referral, and follow-up of patients with or at risk of developing early ILDs?

STUDY DESIGN AND METHODS

Pulmonologists and radiologists with expertise in ILD participated in two iterative rounds of surveys. The surveys aimed to establish consensus regarding ILA reporting, identification of patients with ILA, and identification of populations that might benefit from screening for ILD. Recommended referral criteria and follow-up processes were also addressed. Threshold for consensus was defined a priori as ≥ 75% agreement or disagreement.

RESULTS

Fifty-five experts were invited and 44 participated; consensus was reached on 39 of 85 questions. The following clinically important statements achieved consensus: honeycombing and traction bronchiectasis or bronchiolectasis indicate potentially progressive ILD; honeycombing detected during lung cancer screening should be reported as potentially significant (eg, with the Lung CT Screening Reporting and Data System "S-modifier" [Lung-RADS; which indicates clinically significant or potentially significant noncancer findings]), recommending referral to a pulmonologist in the radiology report; high-resolution CT imaging and full pulmonary function tests should be ordered if nondependent subpleural reticulation, traction bronchiectasis, honeycombing, centrilobular ground-glass nodules, or patchy ground-glass opacity are observed on CT imaging; patients with honeycombing or traction bronchiectasis should be referred to a pulmonologist irrespective of diffusion capacity values; and patients with systemic sclerosis should be screened with pulmonary function tests for early-stage ILD.

INTERPRETATION

Guidance was established for identifying clinically relevant ILA, subsequent referral, and follow-up. These results lay the foundation for developing practical guidance on managing patients with ILA.

Diagnosis and monitoring of systemic sclerosis-associated interstitial lung disease using high-resolution computed tomography

Patients with systemic sclerosis are at high risk of developing systemic sclerosis–associated interstitial lung disease. Symptoms and outcomes of systemic sclerosis–associated interstitial lung disease range from subclinical lung involvement to respiratory failure and death. Early and accurate diagnosis of systemic sclerosis–associated interstitial lung disease is therefore important to enable appropriate intervention. The most sensitive and specific way to diagnose systemic sclerosis–associated interstitial lung disease is by high-resolution computed tomography, and experts recommend that high-resolution computed tomography should be performed in all patients with systemic sclerosis at the time of initial diagnosis. In addition to being an important screening and diagnostic tool, high-resolution computed tomography can be used to evaluate disease extent in systemic sclerosis–associated interstitial lung disease and may be helpful in assessing prognosis in some patients. Currently, there is no consensus with regards to frequency and scanning intervals in patients at risk of interstitial lung disease development and/or progression. However, expert guidance does suggest that frequency of screening using high-resolution computed tomography should be guided by risk of developing interstitial lung disease. Most experienced clinicians would not repeat high-resolution computed tomography more than once a year or every other year for the first few years unless symptoms arose. Several computed tomography techniques have been developed in recent years that are suitable for regular monitoring, including low-radiation protocols, which, together with other technologies, such as lung ultrasound and magnetic resonance imaging, may further assist in the evaluation and monitoring of patients with systemic sclerosis–associated interstitial lung disease. A video abstract to accompany this article is available at: https://www.globalmedcomms.com/respiratory/Khanna/HRCTinSScILD

Phase 2 trial design of BMS-986278, a lysophosphatidic acid receptor 1 (LPA1) antagonist, in patients with idiopathic pulmonary fibrosis (IPF) or progressive fibrotic interstitial lung disease (PF-ILD)

Introduction

Idiopathic pulmonary fibrosis (IPF) and non-IPF, progressive fibrotic interstitial lung diseases (PF-ILD), are associated with a progressive loss of lung function and a poor prognosis. Treatment with antifibrotic agents can slow, but not halt, disease progression, and treatment discontinuation because of adverse events is common. Fibrotic diseases such as these can be mediated by lysophosphatidic acid (LPA), which signals via six LPA receptors (LPA_1–6). Signalling via LPA₁ appears to be fundamental in the pathogenesis of fibrotic diseases. BMS-986278, a second-generation LPA₁ antagonist, is currently in phase 2 development as a therapy for IPF and PF-ILD.

Methods and analysis

This phase 2, randomised, double-blind, placebo-controlled, parallel-group, international trial will include adults with IPF or PF-ILD. The trial will consist of a 42-day screening period, a 26-week placebo-controlled treatment period, an optional 26-week active-treatment extension period, and a 28-day post-treatment follow-up. Patients in both the IPF (n=240) and PF-ILD (n=120) cohorts will be randomised 1:1:1 to receive 30 mg or 60 mg BMS-986278, or placebo, administered orally two times per day for 26 weeks in the placebo-controlled treatment period. The primary endpoint is rate of change in per cent predicted forced vital capacity from baseline to week 26 in the IPF cohort.

Ethics and dissemination

This study will be conducted in accordance with Good Clinical Practice guidelines, Declaration of Helsinki principles, and local ethical and legal requirements. Results will be reported in a peer-reviewed publication.

Trial registration number

NCT04308681.

Radiographic read paradigms and the roles of the central imaging laboratory in neuro-oncology clinical trials

Determination of therapeutic benefit in intracranial tumors is intimately dependent on serial assessment of radiographic images. The Response Assessment in Neuro-Oncology (RANO) criteria were established in 2010 to provide an updated framework to better characterize tumor response to contemporary treatments. Since this initial update a number of RANO criteria have provided some basic principles for the interpretation of changes on MR images; however, the details of how to operationalize RANO and other criteria for use in clinical trials are ambiguous and not standardized. In this review article designed for the neuro-oncologist or treating clinician, we outline essential steps for performing radiographic assessments by highlighting primary features of the Imaging Charter (referred to as the Charter for the remainder of this article), a document that describes the clinical trial imaging methodology and methods to ensure operationalization of the Charter into the workings of a clinical trial. Lastly, we provide recommendations for specific changes to optimize this methodology for neuro-oncology, including image registration, requirement of growing tumor for eligibility in trials of recurrent tumor, standardized image acquisition guidelines, and hybrid reader paradigms that allow for both unbiased measurements and more comprehensive interpretation.

MGA-NET: MULTI-SCALE GUIDED ATTENTION MODELS FOR AN AUTOMATED DIAGNOSIS OF IDIOPATHIC PULMONARY FIBROSIS (IPF)

We propose a Multi-scale, domain knowledge-Guided Attention model (MGA-Net) for a weakly supervised problem - disease diagnosis with only coarse scan-level labels. The use of guided attention models encourages the deep learning-based diagnosis model to focus on the area of interests (in our case, lung parenchyma), at different resolutions, in an end-to-end manner. The research interest is to diagnose subjects with idiopathic pulmonary fibrosis (IPF) among subjects with interstitial lung disease (ILD) using an axial chest high resolution computed tomography (HRCT) scan. Our dataset contains 279 IPF patients and 423 non-IPF ILD patients. The network's performance was evaluated by the area under the receiver operating characteristic curve (AUC) with standard errors (SE) using stratified five-fold cross validation. We observe that without attention modules, the IPF diagnosis model performs unsatisfactorily (AUC±SE =0.690 ± 0.194); by including unguided attention module, the IPF diagnosis model reaches satisfactory performance (AUC±SE =0.956±0.040), but lack explainability; when including only guided high- or medium- resolution attention, the learned attention maps highlight the lung areas but the AUC decreases; when including both high- and medium- resolution attention, the model reaches the highest AUC among all experiments (AUC± SE =0.971 ±0.021) and the estimated attention maps concentrate on the regions of interests for this task. Our results suggest that, for a weakly supervised task, MGA-Net can utilize the population-level domain knowledge to guide the training of the network in an end-to-end manner, which increases both model accuracy and explainability.

End-to-end domain knowledge-assisted automatic diagnosis of idiopathic pulmonary fibrosis (IPF) using computed tomography (CT)

PURPOSE

Domain knowledge (DK) acquired from prior studies is important for medical diagnosis. This paper leverages the population-level DK using an optimality design criterion to train a deep learning model in an end-to-end manner. In this study, the problem of interest is at the patient level to diagnose a subject with idiopathic pulmonary fibrosis (IPF) among subjects with interstitial lung disease (ILD) using a computed tomography (CT). IPF diagnosis is a complicated process with multidisciplinary discussion with experts and is subject to interobserver variability, even for experienced radiologists. To this end, we propose a new statistical method to construct a time/memory-efficient IPF diagnosis model using axial chest CT and DK, along with an optimality design criterion via a DK-enhanced loss function of deep learning.

METHODS

Four state-of-the-art two-dimensional convolutional neural network (2D-CNN) architectures (MobileNet, VGG16, ResNet-50, and DenseNet-121) and one baseline 2D-CNN are implemented to automatically diagnose IPF among ILD patients. Axial lung CT images are retrospectively acquired from 389 IPF patients and 700 non-IPF ILD patients in five multicenter clinical trials. To enrich the sample size and boost model performance, we sample 20 three-slice samples (triplets) from each CT scan, where these three slices are randomly selected from the top, middle, and bottom of both lungs respectively. Model performance is evaluated using a fivefold cross-validation, where each fold was stratified using a fixed proportion of IPF vs non-IPF.

RESULTS

Using DK-enhanced loss function increases the model performance of the baseline CNN model from 0.77 to 0.89 in terms of study-wise accuracy. Four other well-developed models reach satisfactory model performance with an overall accuracy >0.95 but the benefits brought on by the DK-enhanced loss function is not noticeable.

CONCLUSIONS

We believe this is the first attempt that (a) uses population-level DK with an optimal design criterion to train deep learning-based diagnostic models in an end-to-end manner and (b) focuses on patient-level IPF diagnosis. Further evaluation of using population-level DK on prospective studies is warranted and is underway.

Large-Scale Characterization of Systemic Sclerosis Serum Protein Profile: Comparison to Peripheral Blood Cell Transcriptome and Correlations With Skin/Lung Fibrosis

OBJECTIVE

To provide a large-scale assessment of serum protein dysregulation in diffuse cutaneous systemic sclerosis (dcSSc) and to investigate serum protein correlates of SSc fibrotic features.

METHODS

We investigated serum protein profiles of 66 participants with dcSSc at baseline who were enrolled in the Scleroderma: Cyclophosphamide or Transplant Trial and 66 age- and sex-matched healthy control subjects. A panel of 230 proteins, including several cytokines and chemokines, was investigated. Whole blood gene expression profiling in concomitantly collected samples was performed.

RESULTS

Among the participants with dcSSc, the mean disease duration was 2.3 years. All had interstitial lung disease (ILD), and none were being treated with immunosuppressive agents at baseline. Ninety proteins were differentially expressed in participants with dcSSc compared to healthy control subjects. Similar to previous global skin transcript results, hepatic fibrosis, granulocyte and agranulocyte adhesion, and diapedesis were the top overrepresented pathways. Eighteen proteins correlated with the modified Rodnan skin thickness score (MRSS). Soluble epidermal growth factor receptor was significantly down-regulated in dcSSc and showed the strongest negative correlation with the MRSS, being predictive of the score's course over time, whereas α₁ -antichymotrypsin was significantly up-regulated in dcSSc and showed the strongest positive correlation with the MRSS. Furthermore, higher levels of cancer antigen 15-3 correlated with more severe ILD, based on findings of reduced forced vital capacity and higher scores of disease activity on high-resolution computed tomography. Only 14 genes showed significant differential expression in the same direction in serum protein and whole blood RNA gene expression analyses.

CONCLUSION

Diffuse cutaneous SSc has a distinct serum protein profile with prominent dysregulation of proteins related to fibrosis and immune cell adhesion/diapedesis. The differential expression for most serum proteins in SSc is likely to originate outside the peripheral blood cells.

The value of imaging and clinical outcomes in a phase II clinical trial of a lysophosphatidic acid receptor antagonist in idiopathic pulmonary fibrosis

BACKGROUND

Idiopathic pulmonary fibrosis (IPF) is a chronic, progressive fibrotic lung disease characterized by worsening dyspnea and lung function and has a median survival of 2-3 years. Forced vital capacity (FVC) is the primary endpoint used most commonly in IPF clinical trials as it is the best surrogate for mortality. This study assessed quantitative scores from high-resolution computed tomography (HRCT) developed by machine learning as a secondary efficacy endpoint in a 26-week phase II study of BMS-986020 - an LPA1 receptor antagonist - in patients with IPF.

METHODS

HRCT scans from 96% (137/142) of randomized subjects were utilized. Quantitative lung fibrosis (QLF) scores were calculated from the HRCT images. QLF improvement was defined as ⩾2% reduction in QLF score from baseline to week 26.

RESULTS

In the placebo arm, 5% of patients demonstrated an improvement in QLF score at week 26 compared with 15% and 27% of patients in the BMS-986020 600 mg once daily (QD) and twice daily (BID) arms, respectively [versus placebo: p = 0.08 (600 mg QD); p = 0.0098 (600 mg BID)]. Significant correlations were found between changes in QLF and changes in percent predicted FVC, diffusing capacity for carbon monoxide (DLCO), and shortness of breath at week 26 (ρ = -0.41, ρ = -0.22, and ρ = 0.27, respectively; all p < 0.01).

CONCLUSIONS

This study demonstrated the utility of quantitative HRCT as an efficacy endpoint for IPF in a double-blind, placebo-controlled clinical trial setting.The reviews of this paper are available via the supplemental material section.

Intravenous stem cell dose and changes in quantitative lung fibrosis and DLCO in the AETHER trial: a pilot study

OBJECTIVE

Our purpose was to compare quantitative CT-derived changes in lung fibrosis with pulmonary function, including DLCO, in human subjects with idiopathic pulmonary fibrosis who received an injection of one of two different intravenous doses of human bone-marrow-derived mesenchymal stem cells.

PATIENTS AND METHODS

Two three-subject cohorts from the AETHER trial (Allogeneic Human Cells in subjects with Idiopathic Pulmonary Fibrosis via Intravenous Delivery) underwent high-resolution CT and clinical testing at baseline, 24 weeks, and 48 weeks after injection. Cohort 1 received 2x107 stem cells, and cohort 2 received 1x108 stem cells. CT scans were quantitatively analyzed for lung fibrosis using 510K cleared validated software. The percent predicted DLCO and other pulmonary function studies were obtained.

RESULTS

The cohorts were well matched in lung fibrosis at baseline as assessed by CT scan and lung function. The mean QLF in cohort 1 increased from 13.1% at baseline to 17.1% at 48 weeks, while mean QLF in cohort 2 increased from 15.4% at baseline to 16.5% at 48 weeks. The subjects in cohort 2 progressed more slowly in whole lung fibrosis by a mean of 2.87% compared with cohort 1 (p=0.001 with adjustment of baseline covariates) during the baseline to the 48-week interval. The baseline DLCO was lower in cohort 2 than in cohort 1 (p<0.0001). Over 48 weeks of the study, cohort 2 subjects demonstrated a mean DLCO decline of only 2% compared with a decline of 17% in cohort 1 subjects (p=0.02).

CONCLUSIONS

In this pilot study, the subjects receiving 1x108 stem cells demonstrated slower progression in quantitative lung fibrosis and a smaller decrease in DLCO than subjects receiving 2x107 stem cells.

High throughput image labeling on chest computed tomography by deep learning

When mining image data from PACs or clinical trials or processing large volumes of data without curation, the relevant scans must be identified among irrelevant or redundant data. Only images acquired with appropriate technical factors, patient positioning, and physiological conditions may be applicable to a particular image processing or machine learning task. Automatic labeling is important to make big data mining practical by replacing conventional manual review of every single-image series. Digital imaging and communications in medicine headers usually do not provide all the necessary labels and are sometimes incorrect. We propose an image-based high throughput labeling pipeline using deep learning, aimed at identifying scan direction, scan posture, lung coverage, contrast usage, and breath-hold types. They were posed as different classification problems and some of them involved further segmentation and identification of anatomic landmarks. Images of different view planes were used depending on the specific classification problem. All of our models achieved accuracy > 99 % on test set across different tasks using a research database from multicenter clinical trials.

Using Transitional Changes on High-Resolution Computed Tomography to Monitor the Impact of Cyclophosphamide or Mycophenolate Mofetil on Systemic Sclerosis-Related Interstitial Lung Disease

OBJECTIVE

To examine changes in the extent of specific patterns of interstitial lung disease (ILD) as they transition from one pattern to another in response to immunosuppressive therapy in systemic sclerosis-related ILD (SSc-ILD).

METHODS

We evaluated changes in the quantitative extent of specific lung patterns of ILD using volumetric high-resolution computed tomography (HRCT) scans obtained at baseline and after 2 years of therapy in patients treated with either cyclophosphamide (CYC) for 1 year or mycophenolate mofetil (MMF) for 2 years in Scleroderma Lung Study II. ILD patterns included lung fibrosis, ground glass, honeycombing, and normal lung. Net change was calculated as the difference in the probability of change from one ILD pattern to another. Wilcoxon's signed rank test was used to compare the changes.

RESULTS

Forty-seven and 50 patients had baseline and follow-up scans in the CYC and MMF groups, respectively. Mean net improvements reflecting favorable changes from one ILD pattern to another in the whole lung in the CYC and MMF groups, respectively, were as follows: from lung fibrosis to a normal lung pattern, 21% and 19%; from a ground-glass pattern to a normal lung pattern, 30% and 28%; and from lung fibrosis to a ground-glass pattern, 5% and 0.5%. The mean overall improvement in transitioning from a ground-glass pattern or lung fibrosis to a normal lung pattern was significant for both treatments (all P < 0.001).

CONCLUSION

Significantly favorable transitions from both ground-glass and lung fibrosis ILD patterns to a normal lung pattern were observed in patients undergoing immunosuppressive treatment for SSc-ILD, suggesting the usefulness of examining these transitions for insights into the underlying pathobiology of treatment response.

Prediction of progression in idiopathic pulmonary fibrosis using CT scans at baseline: A quantum particle swarm optimization - Random forest approach

Idiopathic pulmonary fibrosis (IPF) is a fatal lung disease characterized by an unpredictable progressive decline in lung function. Natural history of IPF is unknown and the prediction of disease progression at the time of diagnosis is notoriously difficult. High resolution computed tomography (HRCT) has been used for the diagnosis of IPF, but not generally for monitoring purpose. The objective of this work is to develop a novel predictive model for the radiological progression pattern at voxel-wise level using only baseline HRCT scans. Mainly, there are two challenges: (a) obtaining a data set of features for region of interest (ROI) on baseline HRCT scans and their follow-up status; and (b) simultaneously selecting important features from high-dimensional space, and optimizing the prediction performance. We resolved the first challenge by implementing a study design and having an expert radiologist contour ROIs at baseline scans, depending on its progression status in follow-up visits. For the second challenge, we integrated the feature selection with prediction by developing an algorithm using a wrapper method that combines quantum particle swarm optimization to select a small number of features with random forest to classify early patterns of progression. We applied our proposed algorithm to analyze anonymized HRCT images from 50 IPF subjects from a multi-center clinical trial. We showed that it yields a parsimonious model with 81.8% sensitivity, 82.2% specificity and an overall accuracy rate of 82.1% at the ROI level. These results are superior to other popular feature selections and classification methods, in that our method produces higher accuracy in prediction of progression and more balanced sensitivity and specificity with a smaller number of selected features. Our work is the first approach to show that it is possible to use only baseline HRCT scans to predict progressive ROIs at 6 months to 1year follow-ups using artificial intelligence.

Prediction of idiopathic pulmonary fibrosis progression using early quantitative changes on CT imaging for a short term of clinical 18-24-month follow-ups

OBJECTIVE

High-resolution computed tomography (HRCT) plays an indispensable role in the diagnosis of idiopathic pulmonary fibrosis (IPF). Due to unpredictability in progression and the short median survival of 2-5 years, it is critical to delineate the patients with rapid progression. The aim is to evaluate the predictability of IPF progression using the early quantitative changes.

METHODS

Automated texture-based quantitative lung fibrosis (QLF) was calculated from the anonymized HRCT. Two datasets were collected retrospectively: (1) a pilot study of 35 subjects with three sequential scans (baseline and 6 and 12 months) to obtain a threshold, where visual assessments were stable at 6 months but worsened at 12 months; (2) 157 independent subjects to test the threshold. Landmark Cox regressions were used to compare the progression-free survival (PFS) defined by pulmonary function using the threshold from the early changes in QLF. C-indexes were reported as estimations of the concordance of prediction.

RESULTS

A threshold of 4% QLF change at 6 months corresponded to the mean change that worsened on HRCT visually at 12 months from the pilot study. Using the threshold, significant differences were found in the independent dataset (hazard ratio (HZ) = 5.92, p = 0.001 by Cox model, C-index = 0.71 at the most severe lobe; and HZ = 3.22, p = 0.012, C-index = 0.68 in the whole lung). Median PFS was 11.9 months for subjects with ≥ 4% changes, whereas median PFS was greater than 18 months for subjects with < 4% changes at the most severe lobe.

CONCLUSION

Early structural changes on HRCT using a quantitative score can predict progression in lung function.

KEY POINTS

• Changes on HRCT using quantitative texture-based scores can play a pivotal role for providing information and an aid tool for timely management decision for patients with IPF. • Quantitative changes on HRCT of 4% or more, which matched 6-month prior changes with visual assessment of worsening, can play a pivotal role for providing prediction of clinical progression by 3-5 folds higher in the next incidence, compared with those of subjects with less than 4% changes. • Early structural changes of 4% or more in a paired HRCT scans derived by quantitative scores can predict the progression in lung function in 1-2 years in subjects with IPF, which is critical information for timely management decision for subjects with IPF where the median survival is 2 to 5 years.

Building a high-resolution T2-weighted MR-based probabilistic model of tumor occurrence in the prostate

PURPOSE

We present a method for generating a T2 MR-based probabilistic model of tumor occurrence in the prostate to guide the selection of anatomical sites for targeted biopsies and serve as a diagnostic tool to aid radiological evaluation of prostate cancer.

MATERIALS AND METHODS

In our study, the prostate and any radiological findings within were segmented retrospectively on 3D T2-weighted MR images of 266 subjects who underwent radical prostatectomy. Subsequent histopathological analysis determined both the ground truth and the Gleason grade of the tumors. A randomly chosen subset of 19 subjects was used to generate a multi-subject-derived prostate template. Subsequently, a cascading registration algorithm involving both affine and non-rigid B-spline transforms was used to register the prostate of every subject to the template. Corresponding transformation of radiological findings yielded a population-based probabilistic model of tumor occurrence. The quality of our probabilistic model building approach was statistically evaluated by measuring the proportion of correct placements of tumors in the prostate template, i.e., the number of tumors that maintained their anatomical location within the prostate after their transformation into the prostate template space.

RESULTS

Probabilistic model built with tumors deemed clinically significant demonstrated a heterogeneous distribution of tumors, with higher likelihood of tumor occurrence at the mid-gland anterior transition zone and the base-to-mid-gland posterior peripheral zones. Of 250 MR lesions analyzed, 248 maintained their original anatomical location with respect to the prostate zones after transformation to the prostate.

CONCLUSION

We present a robust method for generating a probabilistic model of tumor occurrence in the prostate that could aid clinical decision making, such as selection of anatomical sites for MR-guided prostate biopsies.

Quantitative bone scan lesion area as an early surrogate outcome measure indicative of overall survival in metastatic prostate cancer

A clinical validation of the bone scan lesion area (BSLA) as a quantitative imaging biomarker was performed in metastatic castration-resistant prostate cancer (mCRPC). BSLA was computed from whole-body bone scintigraphy at baseline and week 12 posttreatment in a cohort of 198 mCRPC subjects (127 treated and 71 placebo) from a clinical trial involving a different drug from the initial biomarker development. BSLA computation involved automated image normalization, lesion segmentation, and summation of the total area of segmented lesions on bone scan AP and PA views as a measure of tumor burden. As a predictive biomarker, treated subjects with baseline BSLA <200  cm2 had longer survival than those with higher BSLA (HR=0.4 and p<0.001). As a surrogate outcome biomarker, subjects were categorized as progressive disease (PD) if the BSLA increased by a prespecified 30% or more from baseline to week 12 and non-PD otherwise. Overall survival rates between PD and non-PD groups were statistically different (HR=0.64 and p=0.007). Subjects without PD at week 12 had longer survival than subjects with PD: median 398 days versus 280 days. BSLA has now been demonstrated to be an early surrogate outcome for overall survival in different prostate cancer drug treatments.

An Architecture for Computer-Aided Detection and Radiologic Measurement of Lung Nodules in Clinical Trials

Computer tomography (CT) imaging plays an important role in cancer detection and quantitative assessment in clinical trials. High-resolution imaging studies on large cohorts of patients generate vast data sets, which are infeasible to analyze through manual interpretation.

In this article we describe a comprehensive architecture for computer-aided detection (CAD) and surveillance on lung nodules in CT images. Central to this architecture are the analytic components: an automated nodule detection system, nodule tracking capabilities and volume measurement, which are integrated within a data management system that includes mechanisms for receiving and archiving images, a database for storing quantitative nodule measurements and visualization, and reporting tools.

We describe two studies to evaluate CAD technology within this architecture, and the potential application in large clinical trials. The first study involves performance assessment of an automated nodule detection system and its ability to increase radiologist sensitivity when used to provide a second opinion. The second study investigates nodule volume measurements on CT made using a semi-automated technique and shows that volumetric analysis yields significantly different tumor response classifications than a 2D diameter approach. These studies demonstrate the potential of automated CAD tools to assist in quantitative image analysis for clinical trials.

Transitions to different patterns of interstitial lung disease in scleroderma with and without treatment

OBJECTIVES

The aim is to investigate whether the 12-month quantitative changes in high-resolution CT (HRCT) measures of interstitial lung disease (ILD) are different, and to understand how they change, in patients with scleroderma-related ILD who receive drug therapy versus placebo.

METHODS

HRCT images were acquired at baseline and at 12 months in 83 participants in Scleroderma Lung Study I, a clinical trial comparing treatment with oral cyclophosphamide versus placebo. A computer-aided model was used to quantify the extent of fibrotic reticulation, ground glass and honeycomb patterns and quantitative ILD (QILD: sum of these patterns) in the whole lung and the lung zone (upper, middle or lower) of maximal disease involvement.

RESULTS

Mean QILD score decreased by 3.9% in the cyclophosphamide group while increasing by 4.2% in the placebo group in the most severe zone (p=0.01) and decreased by 3.2% in the cyclophosphamide group while increasing by 2.2% in the placebo group in the whole lung (p=0.03). Transitional probabilities demonstrated greater changes from a fibrotic to either a ground glass or normal pattern in the cyclophosphamide group and the reverse in the placebo group.

CONCLUSIONS

Changes in quantitative HRCT measures of ILD provide a sensitive indication of disease progression and response to treatment.

TRIAL REGISTRATION NUMBER

NCT00004563; Post-results.

Ensemble segmentation for GBM brain tumors on MR images using confidence-based averaging

PURPOSE

Ensemble segmentation methods combine the segmentation results of individual methods into a final one, with the goal of achieving greater robustness and accuracy. The goal of this study was to develop an ensemble segmentation framework for glioblastoma multiforme tumors on single-channel T1w postcontrast magnetic resonance images.

METHODS

Three base methods were evaluated in the framework: fuzzy connectedness, GrowCut, and voxel classification using support vector machine. A confidence map averaging (CMA) method was used as the ensemble rule.

RESULTS

The performance is evaluated on a comprehensive dataset of 46 cases including different tumor appearances. The accuracy of the segmentation result was evaluated using the F1-measure between the semiautomated segmentation result and the ground truth.

CONCLUSIONS

The results showed that the CMA ensemble result statistically approximates the best segmentation result of all the base methods for each case.

Changes in right heart haemodynamics and echocardiographic function in an advanced phenotype of pulmonary hypertension and right heart dysfunction associated with pulmonary fibrosis

BACKGROUND

Pulmonary hypertension (PH)-targeted therapy in the setting of pulmonary fibrosis (PF) is controversial; the main clinical concern is worsening of systemic hypoxaemia. We sought to determine the effects of gentle initiation and chronic administration of parenteral treprostinil on right heart function in patients with PF associated with an advanced PH phenotype.

METHODS

Open-label, prospective analysis of patients with PF-PH referred for lung transplantation (LT). Advanced PH was defined as mean pulmonary artery pressure (mPAP) ≥35 mm Hg. We compared haemodynamics, Doppler echocardiography (DE), oxygenation, dyspnoea and quality of life indices, and 6 min walk distance (6MWD) before and 12 weeks after parenteral treprostinil.

RESULTS

15 patients were recruited in the study. After therapy, there were significant improvements in right heart haemodynamics (right atrial pressure (9.5 ± 3.4 vs 6.0 ± 3.7); mPAP (47 ± 8 vs 38.9 ± 13.4); CI (2.3 ± 0.5 vs 2.7 ± 0.6); pulmonary vascular resistance (698 ± 278 vs 496 ± 229); transpulmonary gradient (34.7 ± 8.7 vs 28.5 ± 10.3); mvO2 (65 ± 7.2 vs 70.9 ± 7.4); and stroke volume index (29.2 ± 6.7 vs 33 ± 7.3)) and DE parameters reflecting right heart function (right ventricular (RV) end diastolic area (36.4 ± 5.2 vs 30.9 ± 8.2 cm(2)), left ventricular eccentricity index (1.7 ± 0.6 vs 1.3 ± 0.5), tricuspid annular planar systolic excursion (1.6 ± 0.5 vs 1.9 ± 0.2 cm)). These changes occurred without significant alteration in systemic oxygenation, heart rate, or mean systemic arterial pressure. In addition, improvements were seen in 6MWD (171 ± 93 vs 230 ± 114), 36-Item Short Form Health Survey Mental Component Summary aggregate (38 ± 11 vs 44.2 ± 10.7), University of California, San Diego Shortness of Breath Questionnaire (87 ± 17.1 vs 73.1 ± 21), and brain natriuretic peptide (558 ± 859 vs 228 ± 340).

CONCLUSIONS

PH-targeted therapy may improve right heart haemodynamics and echocardiographic function without affecting systemic oxygen saturation in an advanced PH phenotype associated with RV dysfunction in the setting of PF.

A dose-ranging study of cabozantinib in men with castration-resistant prostate cancer and bone metastases

BACKGROUND

Cabozantinib is an oral MET/VEGFR2 inhibitor. A recent phase II study of cabozantinib (100 mg daily) showed improved bone scans in subjects with metastatic castration-resistant prostate cancer (mCRPC), but adverse events (AE) caused frequent dose reductions. This study was designed to determine the efficacy and tolerability of cabozantinib at lower starting doses.

EXPERIMENTAL DESIGN

An adaptive design was used to determine the lowest active daily dose among 60, 40, and 20 mg. The primary endpoint was week 6 bone scan response, defined as ≥30% decrease in bone scan lesion area. The secondary endpoint was change in circulating tumor cells (CTC).

RESULTS

Among 11 evaluable subjects enrolled at 40 mg, there were 9 partial responses (PR), 1 complete response, and 1 stable disease (SD). Of 10 subjects subsequently enrolled at 20 mg, there were 1 PR, 5 SDs, and 4 with progressive disease. Among 13 subjects enrolled on the 40 mg expansion cohort, there were 6 PRs and 7 SDs. No subjects required dose reduction or treatment interruption at 6 or 12 weeks; 3 subjects at dose level 0 discontinued due to AEs by 12 weeks. At 40 mg, median treatment duration was 27 weeks. 58% of subjects with ≥5 CTCs/7.5 mL at baseline converted to <5.

CONCLUSIONS

Cabozantinib 40 mg daily was associated with a high rate of bone scan response. Cabozantinib 40 mg daily was associated with better tolerability than previously reported for cabozantinib 100 mg daily. These observations informed the design of phase III studies of cabozantinib in mCRPC.

Investigator-sponsored trial of efficacy and tolerability of cabozantinib (cabo) at lower dose: A dose-finding study in men with castration-resistant prostate cancer (CRPC) and bone metastases

4566

Background: Cabo (XL184) is an oral inhibitor of MET and VEGFR2. In a randomized discontinuation trial of cabo 100mg daily, 76% of men with CRPC and bone metastases had partial or complete resolution of bone scan lesions as early as week (wk) 6. However, treatment was limited by adverse events (AEs), with dose reductions in 51% of patients (pts), and discontinuations in 10%. The current study was designed to determine the efficacy and tolerability of cabo at lower starting doses. Methods: An adaptive response scheme was used to determine the lowest active daily cabo dose among dose levels +1 (60mg), 0 (40mg), and -1 (20mg). The primary endpoint was wk 6 bone scan response (BSR) assessed with an automated FDA 510(k) approved computer-aided detection system. A ≥30% decrease in total bone scan lesion area (BSLA) was defined as a response. The first cohort was treated at dose level 0. The number of responses (≥8 vs. <8 among 11 evaluable pts) was used to select the dose level (-1 vs. +1) for the second cohort. Based on the observed BSR rate in the second cohort of 11 pts, a dose was selected for expansion to treat 13 more pts. Results: The study completed planned enrollment of 36 pts. Median age was 66; 44% were docetaxel-pretreated. Among 12 pts enrolled at dose level 0, there were 10 BSRs at wk 6 including 1 complete response (CR), and 1 pt with stable disease (SD). The median decrease in BSLA was 62%. Ten pts evaluated at wk 12 included 9 BSRs (3 CRs), and 1 sustained SD. Among 11 pts then treated at dose level -1, 10 pts were evaluable at wk 6: 1 BSR, 5 SD, and 4 had progressive disease. No pts in the 2 cohorts required dose reduction or treatment interruption at 12 wks; 1 pt discontinued due to grade 3 AEs (anorexia, fatigue). 6/12 pts with ≥6 months follow-up remain on study. 5/5 pts enrolled at 40mg with CTCs ≥5 per 7.5mL converted to <5. Thirteen pts accrued to the expansion cohort at 40mg daily had confirmed high BSR rate. Conclusions: Cabo 40mg daily achieves a high BSR rate in men with CRPC and bone metastases, and is associated with better tolerability than previously reported for cabo 100mg daily.

Abstract B114: Cabozantinib effects on bone metastasis: Computer-aided quantitative bone scan assessment of prostate cancer treatment response

Background: Cabozantinib is an oral inhibitor of MET and VEGFR2 and at a dose of 100 mg qd has demonstrated high rates of partial or complete bone scan resolution in CRPC patients with bone metastases, as assessed by manual visual reads. No accepted approach exists for the assessment of bone scan response in prostate cancer patients. An FDA 510(k) approved image analysis package has been developed that includes a computer-aided detection (CAD) system to detect pixels associated with bone lesions on bone scan with high accuracy, thus facilitating quantitative assessment of tumor burden with advantages of objectivity and consistency over manual approaches. A change in total Bone Scan Lesion Area of 30% or greater has been previously established as a cutoff point to distinguish responders from non-responders [1].

Methods: Original Digital Imaging and Communications in Medicine (DICOM) format images of baseline and week 6 bone scan assessments were collected and analyzed by the automated CAD system then reviewed by a physician experienced with bone scan interpretation as part of an ongoing clinical study of single agent cabozantinib at a lower starting dose of 40 mg qd in CRPC subjects with evidence of bone metastases. All subjects had undergone standard of care whole body scintigraphy, 2–4 hours post-injection of 20–25 mCi of Tc MDP. All images underwent intensity normalization to a bone scan reference to account for differences in radiotracer dosing levels and scan timing to improve reproducibility. Lesions consistent with metastatic foci of disease were automatically segmented based on anatomic region-specific intensity thresholds. Each segmented image was reviewed by a nuclear medicine physician for removal of any remaining false positive regions (e.g. areas of degenerative joint disease). Quantitative assessment of lesion burden was then determined.

Results: 10 CRPC subjects showing evidence of bone metastasis were treated in an ongoing dose-ranging study with cabozantinib and were evaluable for bone scan response assessment at week 6. Reductions in the following parameters (median and range) were observed at week 6 relative to baseline: total Bone Scan Lesion Area (168.5 cm2 [20, 864] to 54 cm2 [0, 431]; 68% reduction), Bone Scan Lesion Count (34 [3, 74] to 3 [0, 64]; 91% reduction), and mean normalized Bone Scan Lesion Intensity (92.2 [56.5, 155.0] to 57.0 [0, 90.5]; 38% reduction). All 10 subjects had a reduction in Bone Scan Lesion Area, with 9 achieving a bone scan response at the week 6 timepoint as assessed by both the CAD system and visual review by the investigator.

Conclusion: A significant reduction in apparent bone scan abnormality was demonstrated by the automated CAD system. The results demonstrate the potential for objective measurement of cabozantinib treatment effects in bone, laying the foundation for further validation against other clinically relevant outcome measures such as pain and overall survival.

Reference:

1. G. H. Chu, M. S. Brown, H. J. Kim, M. Auerbach, C. Poon, B. Ramakrishna, A. Vidovic, D. W. Gjertson, M. J. Morris, S. M. Larson, J. G. Goldin, H. I. Scher. Initial analytic validation of automated bone scan measures for treatment response assessment in patients with metastatic castration-resistant prostate cancer (CRPC). J Clin Oncol 29: 2011 (suppl; abstr e15174).

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2011 Nov 12-16; San Francisco, CA. Philadelphia (PA): AACR; Mol Cancer Ther 2011;10(11 Suppl):Abstract nr B114.

Predicting treatment outcomes and responder subsets in scleroderma-related interstitial lung disease

OBJECTIVE

To identify baseline characteristics of patients with scleroderma-related interstitial lung disease (SSc-ILD) that could serve as predictors of the most favorable response to 12-month treatment with oral cyclophosphamide (CYC).

METHODS

Regression analyses were retrospectively applied to the Scleroderma Lung Study data in order to identify baseline characteristics that correlated with the absolute change in forced vital capacity (FVC) (% predicted values) and the placebo-adjusted change in % predicted FVC over time (the CYC treatment effect).

RESULTS

Completion of the CYC arm of the Scleroderma Lung Study was associated with a placebo-adjusted improvement in the % predicted FVC of 2.11% at 12 months, which increased to 4.16% when patients were followed up for another 6 months (P=0.014). Multivariate regression analyses identified the maximal severity of reticular infiltrates (assessed as maximum fibrosis scores) on high-resolution computed tomography (HRCT) at baseline, the modified Rodnan skin thickness score (MRSS) at baseline, and the Mahler baseline dyspnea index as independent correlates of treatment response. When patients were stratified on the basis of whether 50% or more of any lung zone was involved by reticular infiltrates on HRCT and/or whether patients exhibited an MRSS of at least 23, a subgroup of patients emerged in whom there was an average CYC treatment effect of 9.81% at 18 months (P<0.001). Conversely, there was no treatment effect (a -0.58% difference) in patients with less severe HRCT findings and a lower MRSS at baseline.

CONCLUSION

A retrospective analysis of the Scleroderma Lung Study data identified the severity of reticular infiltrates on baseline HRCT and the baseline MRSS as patient features that might be predictive of responsiveness to CYC therapy.

Reproducibility of volume and densitometric measures of emphysema on repeat computed tomography with an interval of 1 week

OBJECTIVES

The reproducibilities of CT lung volume and densitometric measures of emphysema were assessed over 1 week. The influence of breathhold on reproducibility was assessed.

METHODS

HRCT was performed on 44 subjects at inspiration on two visits with a 7-day interval. CT lung volume, relative area below -950HU (RA950-raw), and 15th percentile density (PD15-raw) were computed. Volume correction was used to obtain RA950-adj and PD15-adj. Reproducibilities between visits were assessed using concordance correlation coefficient (CCC) and repeatability coefficient (RC). Reproducibilities were compared between raw and adjusted measures. Differences between visits were computed for volume and density measures. Correlations were computed for density differences versus volume difference. Subgroup analysis was performed using a 0.25 L volume difference threshold.

RESULTS

High CCC were observed for all measures in full group (CCC > 0.97). Reproducibilities of volume (RC = 0.67 L), RA950-raw (RC = 2.3%), and PD15-raw (RC = 10.6HU) were observed. Volume correction significantly improved PD15 (RC = 3.6HU) but not RA950 (RC = 1.7%). RA950-raw and PD15-raw had significantly better RC in <0.25 L subgroup than ≥0.25 L. Significant correlations with volume were observed for RA950-raw and PD15-raw (R (2) > 0.71), but not RA950-adj or PD15-adj (R (2) < 0.11).

CONCLUSIONS

Good breathhold and RA950 reproducibilities were achieved. PD15 was less reproducible but improved with volume correction or superior breathhold reproduction.

KEY POINTS

• Good breath-hold reproducibility is achievable between multiple CT examinations. • Reproducibility of densitometric measures may be improved by statistical volume correction. • Volume correction may result in decreased signal. • Densitometric reproducibility may also be improved by achieving good breath-hold reproduction. • Careful consideration of signal and noise is necessary in reproducibility assessment.

A method for the automatic quantification of the completeness of pulmonary fissures: evaluation in a database of subjects with severe emphysema

OBJECTIVES

To propose and evaluate a technique for automatic quantification of fissural completeness from chest computed tomography (CT) in a database of subjects with severe emphysema.

METHODS

Ninety-six CT studies of patients with severe emphysema were included. The lungs, fissures and lobes were automatically segmented. The completeness of the fissures was calculated as the percentage of the lobar border defined by a fissure. The completeness score of the automatic method was compared with a visual consensus read by three radiologists using boxplots, rank sum tests and ROC analysis.

RESULTS

The consensus read found 49% (47/96), 15% (14/96) and 67% (64/96) of the right major, right minor and left major fissures to be complete. For all fissures visually assessed as being complete the automatic method resulted in significantly higher completeness scores (mean 92.78%) than for those assessed as being partial or absent (mean 77.16%; all p values <0.001). The areas under the curves for the automatic fissural completeness were 0.88, 0.91 and 0.83 for the right major, right minor and left major fissures respectively.

CONCLUSIONS

An automatic method is able to quantify fissural completeness in a cohort of subjects with severe emphysema consistent with a visual consensus read of three radiologists.

KEY POINTS

• Lobar fissures are important for assessing the extent and distribution of lung disease • Modern CT allows automatic lobar segmentation and assessment of the fissures • This segmentation can also assess the completeness of the fissures. • Such assessment is important for decisions about novel therapies (eg for emphysema).

Impact and costs of targeted recruitment of minorities to the National Lung Screening Trial

BACKGROUND

To promote results in the National Lung Screening Trial (NLST) that are generalizable across the entire US population, a subset of NLST sites developed dedicated strategies for minority recruitment.

PURPOSE

To report the effects of targeted strategies on the accrual of underrepresented groups, to describe participant characteristics, and to estimate the costs of targeted enrollment.

METHODS

The 2002-2004 Tobacco Use Supplement was used to estimate eligible proportions of racial and ethnic categories. Strategic planning included meetings/conferences with key stakeholders and minority organizations. Potential institutions were selected based upon regional racial/ethnic diversity and proven success in recruitment of underrepresented groups. Seven institutions submitted targeted recruitment strategies with budgets. Accrual by racial/ethnic category was tracked for each institution. Cost estimates were based on itemized receipts for minority strategies relative to minority accrual.

RESULTS

Of 18,842 participants enrolled, 1576 (8.4%) were minority participants. The seven institutions with targeted recruitment strategies accounted for 1223 (77.6%) of all minority participants enrolled. While there was a significant increase in the rate of minority accrual pre-implementation to post-implementation for the institutions with targeted recruitment (9.3% vs. 15.2%, p <  0.0001), there was no significant difference for the institutions without (3.5% vs. 3.8%, p = 0.46). Minority enrollees at the seven institutions tended to have less than a high school education, be economically disadvantaged, and were more often uninsured. These socio-demographic differences persisted at the seven institutions even after adjusting for race and ethnicity. The success of different strategies varied by institution, and no one strategy was successful across all institutions. Costs for implementation were also highly variable, ranging from $146 to $749 per minority enrollee.

LIMITATIONS

Data on minority recruitment processes were not consistently kept at the individual institutions. In addition, participant responses via newspaper advertisements and the efforts of minority staff hired by the institutions could not be coded on Case Report Forms.

CONCLUSIONS

Strategic efforts were associated with significant increases in minority enrollment. The greatest successes require that a priori goals be established based on eligible racial/ethnic proportions; the historical performance of sites in minority accrual should factor into the selection of sites; recruitment planning must begin well in advance of trial launch; and there must be endorsement by prominent representatives of the racial groups of interest.

A computer-aided diagnosis system for quantitative scoring of extent of lung fibrosis in scleroderma patients

OBJECTIVES

To evaluate an improved quantitative lung fibrosis score based on a computer-aided diagnosis (CAD) system that classifies CT pixels with the visual semi-quantitative pulmonary fibrosis score in patients with scleroderma-related interstitial lung disease (SSc-ILD).

METHODS

High-resolution, thin-section CT images were obtained and analysed on 129 subjects with SSc-ILD (36 men, 93 women; mean age 48.8±12.1 years) who underwent baseline CT in the prone position at full inspiration. The CAD system segmented each lung of each patient into 3 zones. A quantitative lung fibrosis (QLF) score was established via 5 steps: 1) images were denoised; 2) images were grid sampled; 3) the characteristics of grid intensities were converted into texture features; 4) texture features classified pixels as fibrotic or non-fibrotic, with fibrosis defined by a reticular pattern with architectural distortion; and 5) fibrotic pixels were reported as percentages. Quantitative scores were obtained from 709 zones with complete data and then compared with ordinal scores from two independent expert radiologists. ROC curve analyses were used to measure performance.

RESULTS

When the two radiologists agreed that fibrosis affected more than 1% or 25% of a zone or zones, the areas under the ROC curves for QLF score were 0.86 and 0.96, respectively.

CONCLUSIONS

Our technique exhibited good accuracy for detecting fibrosis at a threshold of both 1% (i.e. presence or absence of pulmonary fibrosis) and a clinically meaningful threshold of 25% extent of fibrosis in patients with SSc-ILD.

Chronic obstructive pulmonary disease phenotypes: the future of COPD

Significant heterogeneity of clinical presentation and disease progression exists within chronic obstructive pulmonary disease (COPD). Although FEV(1) inadequately describes this heterogeneity, a clear alternative has not emerged. The goal of phenotyping is to identify patient groups with unique prognostic or therapeutic characteristics, but significant variation and confusion surrounds use of the term "phenotype" in COPD. Phenotype classically refers to any observable characteristic of an organism, and up until now, multiple disease characteristics have been termed COPD phenotypes. We, however, propose the following variation on this definition: "a single or combination of disease attributes that describe differences between individuals with COPD as they relate to clinically meaningful outcomes (symptoms, exacerbations, response to therapy, rate of disease progression, or death)." This more focused definition allows for classification of patients into distinct prognostic and therapeutic subgroups for both clinical and research purposes. Ideally, individuals sharing a unique phenotype would also ultimately be determined to have a similar underlying biologic or physiologic mechanism(s) to guide the development of therapy where possible. It follows that any proposed phenotype, whether defined by symptoms, radiography, physiology, or cellular or molecular fingerprint will require an iterative validation process in which "candidate" phenotypes are identified before their relevance to clinical outcome is determined. Although this schema represents an ideal construct, we acknowledge any phenotype may be etiologically heterogeneous and that any one individual may manifest multiple phenotypes. We have much yet to learn, but establishing a common language for future research will facilitate our understanding and management of the complexity implicit to this disease.

Reproducibility of lung and lobar volume measurements using computed tomography

RATIONALE AND OBJECTIVES

Lung and lobar volume measurements from computed tomographic (CT) imaging are being used in clinical trials to assess new minimally invasive emphysema treatments aiming to reduce lung volumes. Establishing the reproducibility of lung volume measurements is important if they are to be accepted as treatment planning and outcome variables. The aims of this study were to (1) investigate the correlation between lung volumes assessed on CT imaging and on pulmonary function testing (PFT), (2) compare the two methods' reproducibility, and (3) assess the reproducibility of CT lobar volumes.

MATERIALS AND METHODS

CT imaging and body plethysmography were performed at baseline and after a 9-month interval in multicenter emphysema treatment trials. Lung volumes were measured at total lung capacity (TLC) and at residual volume (RV). Lobar volumes were measured on CT imaging using a semiautomated technique. The correlations between CT and PFT volumes were computed for 486 subjects at baseline. Reproducibility was assessed in terms of the intraclass correlation coefficient (ICC) for 126 subjects from the control group at TLC and 120 subjects at RV.

RESULTS

Correlations between CT and PFT lung volumes were 0.86 at TLC and 0.67 at RV. At TLC, the ICCs were 0.943 for CT imaging and 0.814 for PFT. At RV, the ICCs were 0.886 for CT imaging and 0.683 for PFT. CT lobar volumes showed good reproducibility (all P values < .05).

CONCLUSION

CT lung and lobar volume measurements could be captured in a multicenter trial setting with high reproducibility and were highly correlated with those obtained on PFT. CT imaging showed significantly better reproducibility than PFT between interval lung volume measurements, offering the potential for designing emphysema treatment trials involving fewer subjects.

Eosinophil and T cell markers predict functional decline in COPD patients

Background

The major marker utilized to monitor COPD patients is forced expiratory volume in one second (FEV1). However, asingle measurement of FEV1 cannot reliably predict subsequent decline. Recent studies indicate that T lymphocytes and eosinophils are important determinants of disease stability in COPD. We therefore measured cytokine levels in the lung lavage fluid and plasma of COPD patients in order to determine if the levels of T cell or eosinophil related cytokines were predictive of the future course of the disease.

Methods

Baseline lung lavage and plasma samples were collected from COPD subjects with moderately severe airway obstruction and emphysematous changes on chest CT. The study participants were former smokers who had not had a disease exacerbation within the past six months or used steroids within the past two months. Those subjects who demonstrated stable disease over the following six months (ΔFEV1 % predicted = 4.7 ± 7.2; N = 34) were retrospectively compared with study participants who experienced a rapid decline in lung function (ΔFEV1 % predicted = -16.0 ± 6.0; N = 16) during the same time period and with normal controls (N = 11). Plasma and lung lavage cytokines were measured from clinical samples using the Luminex multiplex kit which enabled the simultaneous measurement of several T cell and eosinophil related cytokines.

Results and Discussion

Stable COPD participants had significantly higher plasma IL-2 levels compared to participants with rapidly progressive COPD (p = 0.04). In contrast, plasma eotaxin-1 levels were significantly lower in stable COPD subjects compared to normal controls (p < 0.03). In addition, lung lavage eotaxin-1 levels were significantly higher in rapidly progressive COPD participants compared to both normal controls (p < 0.02) and stable COPD participants (p < 0.05).

Conclusion

These findings indicate that IL-2 and eotaxin-1 levels may be important markers of disease stability in advanced emphysema patients. Prospective studies will need to confirm whether measuring IL-2 or eotaxin-1 can identify patients at risk for rapid disease progression.

Update on radiology of emphysema and therapeutic implications

Emerging treatments require appropriate CT targeting of a selected lobe or lobes and target airways to obtain a successful response. CT scan is used in pretreatment planning to select patients and plan treatment strategy and posttreatment to confirm correct deployment of devices and assess treatment response. Increasingly treatments are being developed to treat patients who have emphysema who require accurate quantitation of extent and distribution of the process. Functional assessment can be made by inference of detailed anatomic correlates and by direct measurement of regional function using dynamic scan protocols. This article summarizes the current role of imaging in the assessment of patients who have emphysema.

Computer-aided detection of endobronchial valves using volumetric CT

RATIONALE AND OBJECTIVES

The ability to automatically detect and monitor implanted devices may serve an important role in patient care by aiding the evaluation of device and treatment efficacy. The purpose of this research was to develop a system for the automated detection of one-way endobronchial valves that were implanted for less invasive lung volume reduction.

MATERIALS AND METHODS

Volumetric thin-section computed tomographic data was obtained for 194 subjects; 95 subjects implanted with 246 devices were used for system development and 99 subjects implanted with 354 devices were reserved for testing. The detection process consisted of preprocessing, pattern recognition based detection, and a final device selection. Following the preprocessing, a set of classifiers was trained using AdaBoost to discriminate true devices from false positives. The classifiers in the cascade used two simple features (either the mean or maximum attenuation) of a local region computed at multiple fixed landmarks relative to a template model of the valve.

RESULTS

Free-response receiver-operating characteristic analysis was performed for the evaluation; the system could be set so the mean sensitivity was 96.5% with a mean of 0.18 false positives per subject. If knowledge of the number of implanted devices were incorporated, the sensitivity would be 96.9% with a mean of 0.061 false positives per subject; this corresponds to a total of 12 false negatives and six false positives for the 99 subjects in the test dataset.

CONCLUSION

Software was developed for automated detection of endobronchial valves on volumetric computed tomography. The proposed device modeling and detection techniques may be applicable to other devices as well as useful for evaluation of treatment response.

Monte Carlo simulations to assess the effects of tube current modulation on breast dose for multidetector CT

Tube current modulation was designed to reduce radiation dose in CT imaging while maintaining overall image quality. This study aims to develop a method for evaluating the effects of tube current modulation (TCM) on organ dose in CT exams of actual patient anatomy. This method was validated by simulating a TCM and a fixed tube current chest CT exam on 30 voxelized patient models and estimating the radiation dose to each patient's glandular breast tissue. This new method for estimating organ dose was compared with other conventional estimates of dose reduction. Thirty detailed voxelized models of patient anatomy were created based on image data from female patients who had previously undergone clinically indicated CT scans including the chest area. As an indicator of patient size, the perimeter of the patient was measured on the image containing at least one nipple using a semi-automated technique. The breasts were contoured on each image set by a radiologist and glandular tissue was semi-automatically segmented from this region. Previously validated Monte Carlo models of two multidetector CT scanners were used, taking into account details about the source spectra, filtration, collimation and geometry of the scanner. TCM data were obtained from each patient's clinical scan and factored into the model to simulate the effects of TCM. For each patient model, two exams were simulated: a fixed tube current chest CT and a tube current modulated chest CT. X-ray photons were transported through the anatomy of the voxelized patient models, and radiation dose was tallied in the glandular breast tissue. The resulting doses from the tube current modulated simulations were compared to the results obtained from simulations performed using a fixed mA value. The average radiation dose to the glandular breast tissue from a fixed tube current scan across all patient models was 19 mGy. The average reduction in breast dose using the tube current modulated scan was 17%. Results were size dependent with smaller patients getting better dose reduction (up to 64% reduction) and larger patients getting a smaller reduction, and in some cases the dose actually increased when using tube current modulation (up to 41% increase). The results indicate that radiation dose to glandular breast tissue generally decreases with the use of tube current modulated CT acquisition, but that patient size (and in some cases patient positioning) may affect dose reduction.

Automatic segmentation of lung parenchyma in the presence of diseases based on curvature of ribs

RATIONALE AND OBJECTIVES

Segmentation of lungs using high-resolution computer tomographic images in the setting of diffuse lung diseases is a major challenge in medical image analysis. Threshold-based techniques tend to leave out lung regions that have increased attenuation, such as in the presence of interstitial lung disease. In contrast, streak artifacts can cause the lung segmentation to "leak" into the chest wall. The purpose of this work was to perform segmentation of the lungs using a technique that selects an optimal threshold for a given patient by comparing the curvature of the lung boundary to that of the ribs.

METHODS

Our automated technique goes beyond fixed threshold-based approaches to include lung boundary curvature features. One would expect the curvature of the ribs and the curvature of the lung boundary around the ribs to be very close. Initially, the ribs are segmented by applying a threshold algorithm followed by morphologic operations. The lung segmentation scheme uses a multithreshold iterative approach. The threshold value is verified until the curvature of the ribs and the curvature of the lung boundary are closely matched. The curve of the ribs is represented using polynomial interpolation, and the lung boundary is matched in such a way that there is minimal deviation from this representation. Performance of this technique was compared with conventional (fixed threshold) lung segmentation techniques on 25 subjects using a volumetric overlap fraction measure.

RESULTS

The performance of the rib segmentation technique was significantly different from conventional techniques with an average higher mean volumetric overlap fraction of about 5%.

CONCLUSIONS

The technique described here allows for accurate quantification of volumetric computed tomography and more advanced segmentation of abnormal areas.

Classification of parenchymal abnormality in scleroderma lung using a novel approach to denoise images collected via a multicenter study

RATIONALE AND OBJECTIVES

Computerized classification techniques have been developed to offer accurate and robust pattern recognition in interstitial lung disease using texture features. However, these techniques still present challenges when analyzing computed tomographic (CT) image data from multiprotocols because of disparate acquisition protocols or from standardized, multicenter clinical trials because of noise variability. Our objective is to investigate the utility of denoising thin section CT image data to improve the classification of scleroderma disease patterns. The patterns are lung fibrosis (LF), groundglass (GG), honeycomb (HC), or normal lung (NL) within small regions of interest (ROIs).

METHODS

High-resolution CT images were scanned in a multicenter clinical trial for the Scleroderma Lung Study. A thoracic radiologist contoured a training set (38 patients) consisting of 148 ROIs with 46 LF, 85 GG, 4 HC, and 13 NL patterns and contoured a test set (33 new patients) consisting of 132 ROIs with 44 LF, 72 GG, 4 HC, and 12 NL patterns. The corresponding CT slices of a contoured ROI were denoised using Aujol's mathematic partial differential equation algorithm. The algorithm's noise parameter was estimated as the standard deviation of grey-level signal (in Hounsfield units) in a homogeneous, non-lung region: the aorta. Within each contoured ROI, every pixel within a 4 x 4 neighborhood was sampled (4 x 4 grid sampling). All sampled pixels from a contoured ROI were assumed to be the same disease pattern as labeled by the radiologist. 5,690 pixels (3,009 LF, 1,994 GG, 348 HC, and 339 NL) and 5,045 pixels (2,665 LF, 1,753 GG, 291 HC, and 336 NL) were sampled in training and test sets, respectively. Next, 58 texture features from the original and denoised image were calculated for each pixel. Using a multinomial logistic model, subsets of features (one from original and another from denoised images) were selected to classify disease patterns. Finally, pixels were classified into disease patterns using a support vector machine procedure.

RESULTS

From the training set, multinomial logistic model selected 45 features from the original images and 38 features from denoised images to classify disease patterns. Using the test set, the overall pixel classification rate by SVM increased from 87.8% to 89.5% with denoising. The specific classification rates (original/denoised) were 96.3/96.4% for LF, 88.8/89.4% for GG, 21.3/28.9% for HC, and 73.5/88.4% for NL. Denoising significantly improved the NL and overall classification rates (P = .037 and P = .047 respectively) at ROI level.

CONCLUSIONS

Analyzing multicenter data using a denoising approach led to more parsimonious classification models with increasing accuracy. This approach offers a novel alternate classification strategy for heterogeneous technical and disease components. Furthermore, the model offers the potential to discriminate the multiple patterns of scleroderma disease correctly.

High-resolution CT scan findings in patients with symptomatic scleroderma-related interstitial lung disease

BACKGROUND

Lung disease has become the leading cause of mortality and morbidity in scleroderma (SSc) patients. The frequency, nature, and progression of interstitial lung disease seen on high-resolution CT (HRCT) scans in patients with diffuse SSc (dcSSc) compared with those with limited SSc (lcSSc) has not been well characterized.

METHODS

Baseline HRCT scan images of 162 participants randomized into a National Institutes of Health-funded clinical trial were compared to clinical features, pulmonary function test measures, and BAL fluid cellularity. The extent and distribution of interstitial lung disease HRCT findings, including pure ground-glass opacity (pGGO), pulmonary fibrosis (PF), and honeycomb cysts (HCs), were recorded in the upper, middle, and lower lung zones on baseline and follow-up CT scan studies.

RESULTS

HRCT scan findings included 92.9% PF, 49.4% pGGO, and 37.2% HCs. There was a significantly higher incidence of HCs in the three zones in lcSSc patients compared to dcSSc patients (p = 0.034, p = 0.048, and p = 0.0007, respectively). The extent of PF seen on HRCT scans was significantly negatively correlated with FVC (r = - 0.22), diffusing capacity of the lung for carbon monoxide (r = - 0.44), and total lung capacity (r = - 0.36). A positive correlation was found between pGGO and the increased number of acute inflammatory cells found in BAL fluid (r = 0.28). In the placebo group, disease progression was assessed as 30% in the upper and middle lung zones, and 45% in the lower lung zones. No difference in the progression rate was seen between lcSSc and dcSSc patients.

CONCLUSIONS

PF and GGO were the most common HRCT scan findings in symptomatic SSc patients. HCs were seen in more than one third of cases, being more common in lcSSc vs dcSSc. There was no relationship between progression and baseline PF extent or lcSSc vs dcSSc.

TRIAL REGISTRATION

Clinicaltrials.gov Identifier: NCT00004563.