Synergistic combination of clinical and imaging features predicts abnormal imaging patterns of pulmonary infections
Computers in Biology and Medicine, 43(9), 1241–1251
Abstract
We designed and tested a novel hybrid statistical model that accepts radiologic image features and clinical variables, and integrates this information in order to automatically predict abnormalities in chest computed-tomography (CT) scans and identify potentially important infectious disease biomarkers. In 200 patients, 160 with various pulmonary infections and 40 healthy controls, we extracted 34 clinical variables from laboratory tests and 25 textural features from CT images. From the CT scans, pleural effusion (PE), linear opacity (or thickening) (LT), tree-in-bud (TIB), pulmonary nodules, ground glass opacity (GGO), and consolidation abnormality patterns were analyzed and predicted through clinical, textural (imaging), or combined attributes. The presence and severity of each abnormality pattern was validated by visual analysis of the CT scans. The proposed biomarker identification system included two important steps: (i) a coarse identification of an abnormal imaging pattern by adaptively selected features (AmRMR), and (ii) a fine selection of the most important features from the previous step, and assigning them as biomarkers, depending on the prediction accuracy. Selected biomarkers were used to classify normal and abnormal patterns by using a boosted decision tree (BDT) classifier. For all abnormal imaging patterns, an average prediction accuracy of 76.15% was obtained. Experimental results demonstrated that our proposed biomarker identification approach is promising and may advance the data processing in clinical pulmonary infection research and diagnostic techniques.
Abstract sourceThis 2013 study examined whether clinical laboratory variables and chest CT texture features, used separately or together, could help identify six abnormal imaging patterns associated with pulmonary infections. The retrospective evaluation included 200 patients and healthy controls.
Summary
Bagci and colleagues evaluated a model combining clinical laboratory variables with texture features from chest computed tomography (CT) scans to predict abnormal lung imaging patterns. The study included 200 people: 160 with pulmonary infections and 40 healthy controls. The authors assessed six patterns—pleural effusion, linear opacity or thickening, tree-in-bud, pulmonary nodules, ground-glass opacity, and consolidation—against visual assessments of the scans.
The approach used adaptive feature selection to identify relevant clinical and imaging variables, followed by a boosted decision tree classifier. The paper reports an average prediction accuracy of 76.15% across abnormal imaging patterns. This was a study-specific evaluation, not evidence of prospective clinical validation or readiness for clinical use.
The authors describe the work as a pilot study and note that the dataset had limited variation in clinical features, did not include severe secondary illnesses, and would benefit from larger, standardized, potentially multicenter evaluation.
Cite this work
Ulas Bagci; Kirsten Jaster-Miller; Kenneth N. Olivier; Jianhua Yao; Daniel J. Mollura (2013). Synergistic combination of clinical and imaging features predicts abnormal imaging patterns of pulmonary infections. Computers in Biology and Medicine, 43(9), 1241–1251 https://doi.org/10.1016/j.compbiomed.2013.06.008
BibTeX