Early Diagnosis Of Idiopathic Pulmonary Fibrosis Research Articles

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.

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. 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. 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. 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. 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.

Practical Imaging Interpretation in Patients Suspected of Having Idiopathic Pulmonary Fibrosis: Official Recommendations from the Radiology Working Group of the Pulmonary Fibrosis Foundation.

Imaging serves a key role in the diagnosis of patients suspected of having idiopathic pulmonary fibrosis (IPF). Accurate pattern classification at thin-section chest CT is a key step in multidisciplinary discussions, guiding the need for surgical lung biopsy and determining available pharmacologic therapies. The recent approval of new treatments for fibrosing lung disease has made it more critical than ever for radiologists to facilitate accurate and early diagnosis of IPF. This document was developed by the Radiology Working Group of the Pulmonary Fibrosis Foundation with the goal of providing a practical guide for radiologists. In this review, the critical imaging patterns of IPF, pitfalls in imaging classifications, confounding imaging findings with other fibrotic lung diseases, and reporting standards for cases of lung fibrosis will be discussed.Published under a CC BY 4.0 license.See also the commentary by White and Galvin in this issue.

Cluster analysis based clinical profiling of Idiopathic Pulmonary Fibrosis patients according to comorbidities evident prior to diagnosis: a single-center observational study

BackgroundThe characterization and clinical profiling of people affected by Idiopathic Pulmonary Fibrosis (IPF), based on clinical events occurring prior to the diagnosis of the fibrotic disease, may facilitate the understanding of events and comorbidities that occur before the diagnosis of IPF and aid in identifying patients at an earlier stage of the disease. MethodsIn this observational study, a cohort of 96 patients, obtained from a community-based pulmonary clinic, were studied retrospectively. These patients were diagnosed with IPF between January 2008 and November 2016, based on findings on lung biopsy and/or high-resolution CT. Using clinical data obtained within the five years before diagnosis and the two-step method of cluster analysis, patients were assigned to one of four groups. The distribution of clinical characteristics and comorbidities present prior to diagnosis was analyzed among the clusters. ResultsCluster 1 is composed of male patients, smokers, with ischemic heart disease. Cluster 2 is composed of male patients, smokers with dyspnea at rest, dry cough and prevalent emphysema. Cluster 3 is composed of male patients without other significant comorbidities, half of whom had dyspnea at exertion. Cluster 4 includes female patients only, most of whom never-smokers and the largest portion of patients with hypothyroidism. The majority of patients had basal end-inspiratory crackles at time of diagnosis, equally distributed among clusters. ConclusionDifferent clinical phenotypes of IPF emerge years before time of diagnosis and if confirmed in larger cohorts may help in forming diagnostic algorithms that would allow earlier diagnosis of IPF.

Changes in management of idiopathic pulmonary fibrosis: impact on disease severity and mortality

ABSTRACT Background Idiopathic pulmonary fibrosis (IPF) is a serious interstitial lung disease (ILD) with a median survival of 3-5 years. The aim of the present study was to evaluate disease severity and survival in patients diagnosed with IPF in the era of antifibrotic therapies compared with an earlier IPF cohort. Methods We identified all patients with fibrotic ILD in the hospital electronic case record system between 2011 and 2016, and reviewed each case in order to identify incident patients with IPF. We used the GAP-index to compare disease severity and mortality to previous findings in patients with IPF diagnosed at our center between 2003 and 2009. Results 260 patients were diagnosed with IPF between 2011 and 2016. Mean age was 72.6 years, 79% were male, mean forced vital capacity (FVC) was 80%, and mean diffusing capacity for carbon monoxide (DLco) was 44%. Age, FVC and DLco were significant predictors of mortality, but the presence of a typical usual interstitial pneumonia pattern on HRCT was not. Eighty percent of patients in GAP stage I received antifibrotic therapy, 73% in GAP stage II, and 29% in GAP stage III. The median survival was four years in the 2011-2016 cohort compared with three years in the 2003-2009 cohort. The distribution of patients between GAP stages was unchanged in 2011-2016 compared with 2003-2009, (stage I 34% vs. 32%, stage II 49% vs. 48% and stage III 20% vs. 16%). One-year mortality was 13% in 2011-2016 and 26% in 2003-2009. In severe disease (GAP stage III), one-year mortality was 26% and 54%, respectively, (p=0.019). Conclusion Short-term mortality was significantly lower in the 2011-2016 cohort compared with 2003-2009. This improvement may be linked to changes in treatment strategies towards limited use of corticosteroids. Although early diagnosis of IPF still needs increased focus, the improvement is encouraging.

Idiopathic Pulmonary Fibrosis and Antifibrotic Treatments: Focus on Experimental Studies.

- Idiopathic pulmonary fibrosis (IPF) is a progressive fatal disease that up to now has been associated with a poor outcome. Some advances have been made in understanding the multiple interrelated pathogenic pathways underlying IPF. The disease is now believed to result from complex interactions among genetic, epigenetic, transcriptional, posttranscriptional, metabolic, and environmental factors. The discovery and validation of theranostic biomarkers are necessary to enable a more precise and earlier diagnosis of IPF and to improve the prediction of future disease behavior. Two drugs recently approved by the US Food and Drug Administration, pirfenidone and nintedanib, have shown the ability to reduce the progression of the disease, although survival benefits are only minimal and neither drug prevents or reverses the disease. - To provide a critical overview of the main experimental studies carried out for testing the principal effects of pirfenidone and nintedanib on IPF. - Experimental (animal and in vitro) studies concerning both drugs were used. - Pirfenidone has a longer history of preclinical experimental studies than nintedanib. Many studies have been reported more recently (after 2014) and some of them evaluated the association of both drugs, thus suggesting their combination in future therapeutic approaches. Future investigations focusing on targets at molecular, cellular, and tissue levels are necessary to have a better in-depth knowledge of the properties of these drugs and to explore the potential efficacy of both or other drug combinations.

Current concepts of idiopathic pulmonary fibrosis: focus on biomarkers

Idiopathic pulmonary fibrosis (IPF) is a chronic progressive lung disease with a poor prognosis. In recent years, the development of international diagnostic criteria based on clinical, physiological, radiological, and histopathological appearances, improved diagnosis of IPF. However, currently available clinical data do not accurately predict the course of the disease, which can vary from a slowly progressing to rapidly progressive, and, in 5% of cases, is punctuated by episodes of rapid acute exacerbation. These challenges highlight the need for the development and validation of diagnostic biomarkers specific to IPF and prognostic biomarkers of future disease behavior to guide treatment decisions, including referral for transplant. The recent approval of pirfenidone and nintedanib and the identification of new potential therapeutic targets have created an urgent need for theragnostic markers, i.e. biomarkers able to assess, ideally at an early stage, therapeutic response to a given drug. This will avoid the side effects and increase efficacy of treatment. In addition, the currently available methods are not able to identify the IPF in the early stage to predict the course of disease, and to assess response to antifibrotic therapy. Recent advances in understanding the multiple interrelated pathogenic pathways underlying IPF have identified various molecular phenotypes resulting from complex interactions among genetic, epigenetic, transcriptional, post-transcriptional, metabolic, and environmental factors. The development and validation of diagnostic and prognostic biomarkers are necessary to enable a more precise and earlier diagnosis of IPF and to improve prediction of future disease behavior.

Idiopathic pulmonary fibrosis: current and future directions

Idiopathic pulmonary fibrosis (IPF) is the most common idiopathic interstitial pneumonia and is increasingly recognised. Prior to the advent of effective therapies, achieving an early diagnosis was arguably of little prognostic consequence given IPF was considered an untreatable and uniformly fatal disease. The advent of new drug treatments has given hope for the future and raised the profile of IPF. International management guidelines highlight the critical role of radiology as part of an interstitial lung disease multidisciplinary team approach in reaching an accurate and early diagnosis of IPF. The diagnostic criteria and levels of diagnostic confidence for the radio-pathological pattern associated with the clinical syndrome of IPF, usual interstitial pneumonia (UIP), appear seemingly straightforward; however, with increasing research and recognition of radiopathological interobserver variability, limitations of this classification model are becoming increasingly apparent. This review describes ancillary radiological features, comorbidities, and emerging new entities that potentially co-exist with IPF. Beyond diagnosis radiology is developing as a key prognostic tool to inform longitudinal patient evaluation. These diagnostic and prognostic clinical challenges and the future role of radiology in IPF are discussed.

Recent advances in understanding idiopathic pulmonary fibrosis.

Despite major research efforts leading to the recent approval of pirfenidone and nintedanib, the dismal prognosis of idiopathic pulmonary fibrosis (IPF) remains unchanged. The elaboration of international diagnostic criteria and disease stratification models based on clinical, physiological, radiological, and histopathological features has improved the accuracy of IPF diagnosis and prediction of mortality risk. Nevertheless, given the marked heterogeneity in clinical phenotype and the considerable overlap of IPF with other fibrotic interstitial lung diseases (ILDs), about 10% of cases of pulmonary fibrosis remain unclassifiable. Moreover, currently available tools fail to detect early IPF, predict the highly variable course of the disease, and assess response to antifibrotic drugs. Recent advances in understanding the multiple interrelated pathogenic pathways underlying IPF have identified various molecular phenotypes resulting from complex interactions among genetic, epigenetic, transcriptional, post-transcriptional, metabolic, and environmental factors. These different disease endotypes appear to confer variable susceptibility to the condition, differing risks of rapid progression, and, possibly, altered responses to therapy. The development and validation of diagnostic and prognostic biomarkers are necessary to enable a more precise and earlier diagnosis of IPF and to improve prediction of future disease behaviour. The availability of approved antifibrotic therapies together with potential new drugs currently under evaluation also highlights the need for biomarkers able to predict and assess treatment responsiveness, thereby allowing individualised treatment based on risk of progression and drug response. This approach of disease stratification and personalised medicine is already used in the routine management of many cancers and provides a potential road map for guiding clinical care in IPF.

Neglected evidence in idiopathic pulmonary fibrosis and the importance of early diagnosis and treatment

In idiopathic pulmonary fibrosis (IPF), some facts or concepts based on substantial evidence, whilst implicit for learned subspecialists, have previously been neglected and/or not explicitly formulated or made accessible to a wider audience. IPF is strongly associated with cigarette smoking and is predominantly a disease of ageing. However, its cause(s) remain elusive and, thus, it is one of the most challenging diseases for the development of novel effective and safe therapies. With the approval of pirfenidone for patients with mild-to-moderate IPF, an earlier diagnosis of IPF is a prerequisite for earlier treatment and, potentially, improvement of the long-term clinical outcome of this progressive and ultimately fatal disease. An earlier diagnosis may be achieved in IPF by promoting thin-slice chest high-resolution computed tomography screening of interstitial lung disease as a "by-product" of large-scale lung cancer screening strategies in smokers, but other techniques, which have been neglected in the past, are now available. Lung auscultation and early identification of "velcro" crackles has been proposed as a key component of early diagnosis of IPF. An ongoing study is exploring correlations between lung sounds on auscultation obtained using electronic stethoscopes and high-resolution computed tomography patterns.

Idiopathic pulmonary fibrosis: the need for early diagnosis

Idiopathic pulmonary fibrosis (IPF), a chronic fibrosing lung disease of a progressive nature and unknown etiology, has the largest epidemiological impact and the worst prognosis among the idiopathic interstitial pneumonias (IIP). Despite the progress in pathogenetic knowledge, many aspects are still dubious, in particular the biomolecular mechanisms activated in the early stages of the disease. Early diagnosis is desirable not only to better define aspects of the natural history of the disease, but also to customize treatment protocols. An early diagnosis of IPF should necessarily be based on the ability to highlight a number of features drawn not only from a careful composition of specific anamnestic data with clinical, functional and radiological parameters, but also from biological markers that, in a proper context, can provide guidance and confirm a clinical-anamnestic suspicion. The identification of specific biomarkers for IPF is a modern and attractive look for the potential clinical implications in terms of diagnosis, prediction of disease progression and prognosis. Biomolecular investigations on IPF were performed selectively on tissue samples, bronchoalveolar lavage (BAL), or blood: nowadays the “multi-omic” approach may allow studying individual constitutional profiles resorting to a series of biomolecular disciplines, the so-called “omics”, which focuses on responses of the entire genomic complex, in line with the current trend to quantitatively analyze the interactions of all components of a biological system. Such refined investigations are an essential base for research now, but they might become a routine in the near future, allowing a more precise classification of patients suffering from a disease of unclear taxonomy.

Idiopathic pulmonary fibrosis: the need for early diagnosis

Idiopathic pulmonary fibrosis (IPF), a chronic fibrosing lung disease of a progressive nature and unknown etiology, has the largest epidemiological impact and the worst prognosis among the idiopathic interstitial pneumonias (IIP). Despite the progress in pathogenetic knowledge, many aspects are still dubious, in particular the biomolecular mechanisms activated in the early stages of the disease. Early diagnosis is desirable not only to better define aspects of the natural history of the disease, but also to customize treatment protocols. An early diagnosis of IPF should necessarily be based on the ability to highlight a number of features drawn not only from a careful composition of specific anamnestic data with clinical, functional and radiological parameters, but also from biological markers that, in a proper context, can provide guidance and confirm a clinical-anamnestic suspicion. The identification of specific biomarkers for IPF is a modern and attractive look for the potential clinical implications in terms of diagnosis, prediction of disease progression and prognosis. Biomolecular investigations on IPF were performed selectively on tissue samples, bronchoalveolar lavage (BAL), or blood: nowadays the “multi-omic” approach may allow studying individual constitutional profiles resorting to a series of biomolecular disciplines, the so-called “omics”, which focuses on responses of the entire genomic complex, in line with the current trend to quantitatively analyze the interactions of all components of a biological system. Such refined investigations are an essential base for research now, but they might become a routine in the near future, allowing a more precise classification of patients suffering from a disease of unclear taxonomy.

Idiopathic pulmonary fibrosis in 2011: key updates on guidelines and therapeutics

This first Advancing in Idiopathic pulmonary fibrosis Research (AIR) meeting took place in Berlin, on the 4th and 5th of November 2011, bringing leading experts in the field of Idiopathic Pulmonary Fibrosis (IPF) and European clinicians together to share knowledge on recent advances and practical experience in the management of this devastating disease. It is widely recognised that although IPF is a rare fibrosing lung disease it is associated with a tremendous burden. The estimated median survival of patients with IPF is 2–5 years following diagnosis, and the mortality rate is greater than that associated with numerous malignancies [1,2]. This has presented significant challenges in patient management and until recently there were no pharmacological treatments approved for patients with IPF in Europe. The timing of this meeting was therefore particularly relevant as there were a number of important advances and new findings during 2011, which included the publication of the updated guidelines from the ATS/ERS/JRS/ALAT Committee and the approval of pirfenidone for the treatment of mild-to-moderate IPF in Europe [3,4]. In addition, a press release from the National Institutes of Health reported that one arm of an ongoing trial, PANTHER-IPF, had been stopped because patients with IPF receiving a currently used triple-drug therapy consisting of prednisone, azathioprine, and N-acetylcysteine (NAC) had worse outcomes than those who received matching placebo. The interim results from PANTHER-IPF have since been published [5]. This supplement to Respiratory Research reported the proceedings from this first European IPF meeting, with key articles selected by the AIR Scientific Committee, which were considered essential reading for all clinicians managing patients with IPF in Europe. Articles in this supplement included a review of the recently updated ATS/ERS/JRS ALAT diagnostic criteria by Professor Athol Wells, and discussion on how these can facilitate early diagnosis of IPF. It also considered issues that may remain unclear despite the updated guidance. Linked to the diagnosis of IPF, Nicola Sverzellati gave a master class on HRCT imaging and practical requirements for optimising CT scanning techniques. Presentations then focused on the management of IPF. Professor Luca Richeldi discussed the Cochrane Collaboration approach to analysing clinical trial data and the findings of Cochrane meta-analyses that have been performed to determine the effect of treatments used in the management of IPF. This was followed by Professor Vincent Cottin, who summarised the outcomes from prior clinical trials in IPF and the recently published Phase III studies of pirfenidone that led to its approval in Europe. Following these updates, Professor Jurgen Behr and Luca Richeldi discussed how these data can be interpreted in light of the updated recommendations of the ATS/ERS/JRS/ALAT Committee regarding treatment of IPF. Professor Carlo Albera outlined the implications of the press release from the NIH regarding the PANTHER-IPF study. Hopefully the selected proceedings from this meeting will provide IPF clinicians with new insights and support in the management of their patients. The intention is for the AIR meeting to become a fixed date in the calendar for respiratory physicians and researchers dedicated to improving knowledge and bringing new hope for IPF patients.

Familial pulmonary fibrosis is the strongest risk factor for idiopathic pulmonary fibrosis

Idiopathic pulmonary fibrosis (IPF) is a lethal lung disorder of unknown etiology. The disease is likely the result of complex interactions between genetic and environmental factors. Evidence suggests that certain environmental factors, such as cigarette smoking and metal dust exposures, or comorbidities like gastroesophageal reflux, and type 2 diabetes mellitus (DM2) may increase risk to develop IPF. Substantial uncertainty remains, however, regarding these and other putative risk factors for IPF. In this study we performed a case-control analysis including 100 patients with IPF and 263 controls matched for age sex and place of residence. We used a structured questionnaire to identify potential risk factors for IPF, including environmental and occupational exposures as well as the relevance of family history of pulmonary fibrosis. The multivariate analysis revealed that family history of pulmonary fibrosis [OR = 6.1, CI95% 2.3-15.9; p < 0.0001] was strongly associated with increased risk of IPF. Actually, 20% of the cases reported a parent or sibling with pulmonary fibrosis. Gastroesophageal reflux [OR = 2.9, CI: 1.3-6.6; p = 0.007], former cigarette smoking [OR = 2.5, CI: 1.4-4.6, p = 0.003], and past or current occupational exposure to dusts, smokes, gases or chemicals [OR = 2.8, CI: 1.5-5.5; p = 0.002] were also associated with the disease. Despite being a significant risk factor on univariate analysis DM2 was not significant in multivariate analysis. These findings indicate that family history of pulmonary fibrosis is a strong risk factor for IPF. Also, we confirmed that occupational exposures, gastroesophageal reflux and former smoking increase the risk for this disease.