Adrien Depeursinge1, Anne S Chin, Ann N Leung, Donato Terrone, Michael Bristow, Glenn Rosen, Daniel L Rubin. 1. From the *Department of Radiology, School of Medicine, Stanford University, Stanford, CA; †Institute of Information Systems, University of Applied Sciences Western Switzerland (HES-SO), Sierre, Switzerland; and ‡Section on Biomedical Informatics Research, Department of Medicine, School of Medicine, Stanford University, Stanford, CA.
Abstract
OBJECTIVES: We propose a novel computational approach for the automated classification of classic versus atypical usual interstitial pneumonia (UIP). MATERIALS AND METHODS: Thirty-three patients with UIP were enrolled in this study. They were classified as classic versus atypical UIP by a consensus of 2 thoracic radiologists with more than 15 years of experience using the American Thoracic Society evidence-based guidelines for computed tomography diagnosis of UIP. Two cardiothoracic fellows with 1 year of subspecialty training provided independent readings. The system is based on regional characterization of the morphological tissue properties of lung using volumetric texture analysis of multiple-detector computed tomography images. A simple digital atlas with 36 lung subregions is used to locate texture properties, from which the responses of multidirectional Riesz wavelets are obtained. Machine learning is used to aggregate and to map the regional texture attributes to a simple score that can be used to stratify patients with UIP into classic and atypical subtypes. RESULTS: We compared the predictions on the basis of regional volumetric texture analysis with the ground truth established by expert consensus. The area under the receiver operating characteristic curve of the proposed score was estimated to be 0.81 using a leave-one-patient-out cross-validation, with high specificity for classic UIP. The performance of our automated method was found to be similar to that of the 2 fellows and to the agreement between experienced chest radiologists reported in the literature. However, the errors of our method and the fellows occurred on different cases, which suggests that combining human and computerized evaluations may be synergistic. CONCLUSIONS: Our results are encouraging and suggest that an automated system may be useful in routine clinical practice as a diagnostic aid for identifying patients with complex lung disease such as classic UIP, obviating the need for invasive surgical lung biopsy and its associated risks.
OBJECTIVES: We propose a novel computational approach for the automated classification of classic versus atypical usual interstitial pneumonia (UIP). MATERIALS AND METHODS: Thirty-three patients with UIP were enrolled in this study. They were classified as classic versus atypical UIP by a consensus of 2 thoracic radiologists with more than 15 years of experience using the American Thoracic Society evidence-based guidelines for computed tomography diagnosis of UIP. Two cardiothoracic fellows with 1 year of subspecialty training provided independent readings. The system is based on regional characterization of the morphological tissue properties of lung using volumetric texture analysis of multiple-detector computed tomography images. A simple digital atlas with 36 lung subregions is used to locate texture properties, from which the responses of multidirectional Riesz wavelets are obtained. Machine learning is used to aggregate and to map the regional texture attributes to a simple score that can be used to stratify patients with UIP into classic and atypical subtypes. RESULTS: We compared the predictions on the basis of regional volumetric texture analysis with the ground truth established by expert consensus. The area under the receiver operating characteristic curve of the proposed score was estimated to be 0.81 using a leave-one-patient-out cross-validation, with high specificity for classic UIP. The performance of our automated method was found to be similar to that of the 2 fellows and to the agreement between experienced chest radiologists reported in the literature. However, the errors of our method and the fellows occurred on different cases, which suggests that combining human and computerized evaluations may be synergistic. CONCLUSIONS: Our results are encouraging and suggest that an automated system may be useful in routine clinical practice as a diagnostic aid for identifying patients with complex lung disease such as classic UIP, obviating the need for invasive surgical lung biopsy and its associated risks.
Authors: Joseph P Lynch; Rajan Saggar; S Sam Weigt; David A Zisman; Eric S White Journal: Semin Respir Crit Care Med Date: 2006-12 Impact factor: 3.119
Authors: Deborah Assayag; Brett M Elicker; Thomas H Urbania; Thomas V Colby; Bo Hyoung Kang; Jay H Ryu; Talmadge E King; Harold R Collard; Dong Soon Kim; Joyce S Lee Journal: Radiology Date: 2013-10-28 Impact factor: 11.105
Authors: Esam H Alhamad; Feisal A Al-Kassimi; Ahmad A Alboukai; Emad Raddaoui; Mohammed S Al-Hajjaj; Waseem Hajjar; Shaffi A Shaik Journal: Respir Med Date: 2012-08-05 Impact factor: 3.415
Authors: Baojun Li; Gary E Christensen; Eric A Hoffman; Geoffrey McLennan; Joseph M Reinhardt Journal: Acad Radiol Date: 2012-08-28 Impact factor: 3.173
Authors: Anna J Podolanczuk; Elizabeth C Oelsner; R Graham Barr; Eric A Hoffman; Hilary F Armstrong; John H M Austin; Robert C Basner; Matthew N Bartels; Jason D Christie; Paul L Enright; Bernadette R Gochuico; Karen Hinckley Stukovsky; Joel D Kaufman; P Hrudaya Nath; John D Newell; Scott M Palmer; Dan Rabinowitz; Ganesh Raghu; Jessica L Sell; Jered Sieren; Sushil K Sonavane; Russell P Tracy; Jubal R Watts; Kayleen Williams; Steven M Kawut; David J Lederer Journal: Eur Respir J Date: 2016-07-28 Impact factor: 16.671
Authors: Andreas Christe; Alan A Peters; Dionysios Drakopoulos; Johannes T Heverhagen; Thomas Geiser; Thomai Stathopoulou; Stergios Christodoulidis; Marios Anthimopoulos; Stavroula G Mougiakakou; Lukas Ebner Journal: Invest Radiol Date: 2019-10 Impact factor: 6.016
Authors: Hiram Shaish; Firas S Ahmed; David Lederer; Belinda D'Souza; Paul Armenta; Mary Salvatore; Anjali Saqi; Sophia Huang; Sachin Jambawalikar; Simukayi Mutasa Journal: Ann Am Thorac Soc Date: 2021-01