Predicting Usual Interstitial Pneumonia Histopathology From Chest CT Imaging With Deep Learning

Bratt, Alex; Williams, James Mickel; Liu, Grace; Patel, Parth P; Walkoff, Lara; Sykes, Annemarie; Tandon, Yasmeen K.; François, Christopher J.; Blezek, Daniel J.; Larson, Nicholas B.; Erickson, Bradley J.; Yi, Eunhee S.; Moua, Teng; Koo, Chi Wan

doi:10.1016/j.chest.2022.03.044

Cited by 23 publications

(11 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Shaish et al 48 introduced a CNN that predicted histopathologic UIP on virtual wedges of peripheral lung on HRCT, with sensitivity and specificity of 74% and 58%, respectively, and moderate agreement with expert radiologists. Most recently, using a pathologically proven ILD-trained DL model, Bratt et al 46 showed that DL outperformed visual CT analysis in predicting histopathologic diagnosis of UIP and had greater reproducibility than experts. Specifically, in cases classified by the 2018 ATS/ERS/JRS/ALAT guideline as probable UIP, DL had a specificity of 89% in predicting the histopathology compared with 85% specificity of expert radiologists.…”

Section: Ipfmentioning

confidence: 99%

Recent Advancements in Computed Tomography Assessment of Fibrotic Interstitial Lung Diseases

Suman

Koo

2023

Journal of Thoracic Imaging

Self Cite

View full text Add to dashboard Cite

Interstitial lung disease (ILD) is a heterogeneous group of disorders with complex and varied imaging manifestations and prognosis. High-resolution computed tomography (HRCT) is the current standard-of-care imaging tool for ILD assessment. However, visual evaluation of HRCT is limited by interobserver variation and poor sensitivity for subtle changes. Such challenges have led to tremendous recent research interest in objective and reproducible methods to examine ILDs. Computer-aided CT analysis to include texture analysis and machine learning methods have recently been shown to be viable supplements to traditional visual assessment through improved characterization and quantification of ILDs. These quantitative tools have not only been shown to correlate well with pulmonary function tests and patient outcomes but are also useful in disease diagnosis, surveillance and management. In this review, we provide an overview of recent computer-aided tools in diagnosis, prognosis, and longitudinal evaluation of fibrotic ILDs, while outlining some of the pitfalls and challenges that have precluded further advancement of these tools as well as potential solutions and further endeavors.

show abstract

Section: Ipfmentioning

confidence: 99%

Recent Advancements in Computed Tomography Assessment of Fibrotic Interstitial Lung Diseases

Suman

Koo

2023

Journal of Thoracic Imaging

Self Cite

View full text Add to dashboard Cite

show abstract

“…Deep learning and machine learning involve the creation of algorithms based on previously existing data that can be used to analyse CT scans and formulate a diagnosis. Retrospective analyses found diagnostic accuracy of these methods to be as high as 83.6% 34 and superior to visual analysis in predicting usual interstitial pneumonia (UIP) histology 35 . The application of content‐based image retrieval, a method of retrieving and matching similar images from a database, improves diagnostic accuracy and agreement between radiologists 36 …”

Section: Diagnosis Monitoring and Prognosismentioning

confidence: 99%

Contemporary Concise Review 2022: Interstitial lung disease

Smith

Jenkins

2023

Respirology

View full text Add to dashboard Cite

SUMMARY OF KEY POINTS Novel genetic associations for idiopathic pulmonary fibrosis (IPF) risk have been identified. Common genetic variants associated with IPF are also associated with chronic hypersensitivity pneumonitis. The characterization of underlying mechanisms, such as pathways involved in myofibroblast differentiation, may reveal targets for future treatments. Newly identified circulating biomarkers are associated with disease progression and mortality. Deep learning and machine learning may increase accuracy in the interpretation of CT scans. Novel treatments have shown benefit in phase 2 clinical trials. Hospitalization with COVID‐19 is associated with residual lung abnormalities in a substantial number of patients. Inequalities exist in delivering and accessing interstitial lung disease specialist care.

show abstract

“…Song et al 49 developed a Details Relation Extraction neural network to diagnose COVID-19 from CT images. Bratt et al 50 developed a deep learning method to predict usual interstitial pneumonia histopathology from CT images. Shiri et al 51 developed a deep neural network to assess the severity of COVID-19 based on CT radiomics features.…”

Section: Related Workmentioning

confidence: 99%

Improve the performance of CT-based pneumonia classification via source data reweighting

Xie

Zhao

2023

Sci Rep

View full text Add to dashboard Cite

Pneumonia is a life-threatening disease. Computer tomography (CT) imaging is broadly used for diagnosing pneumonia. To assist radiologists in accurately and efficiently detecting pneumonia from CT scans, many deep learning methods have been developed. These methods require large amounts of annotated CT scans, which are difficult to obtain due to privacy concerns and high annotation costs. To address this problem, we develop a three-level optimization based method which leverages CT data from a source domain to mitigate the lack of labeled CT scans in a target domain. Our method automatically identifies and downweights low-quality source CT data examples which are noisy or have large domain discrepancy with target data, by minimizing the validation loss of a target model trained on reweighted source data. On a target dataset with 2218 CT scans and a source dataset with 349 CT images, our method achieves an F1 score of 91.8% in detecting pneumonia and an F1 score of 92.4% in detecting other types of pneumonia, which are significantly better than those achieved by state-of-the-art baseline methods.

show abstract

Predicting Usual Interstitial Pneumonia Histopathology From Chest CT Imaging With Deep Learning

Cited by 23 publications

References 21 publications

Recent Advancements in Computed Tomography Assessment of Fibrotic Interstitial Lung Diseases

Recent Advancements in Computed Tomography Assessment of Fibrotic Interstitial Lung Diseases

Contemporary Concise Review 2022: Interstitial lung disease

Improve the performance of CT-based pneumonia classification via source data reweighting

Contact Info

Product

Resources

About