Development of a Computer-Aided Differential Diagnosis System to Distinguish Between Usual Interstitial Pneumonia and Non-specific Interstitial Pneumonia Using Texture- and Shape-Based Hierarchical Classifiers on HRCT Images

Abstract

A computer-aided differential diagnosis (CADD) system that distinguishes between usual interstitial pneumonia (UIP) and non-specific interstitial pneumonia (NSIP) using high-resolution computed tomography (HRCT) images was developed, and its results compared against the decision of a radiologist. Six local interstitial lung disease patterns in the images were determined, and 900 typical regions of interest were marked by an experienced radiologist. A support vector machine classifier was used to train and label the regions of interest of the lung parenchyma based on the texture and shape characteristics. Based on the regional classifications of the entire lung using HRCT, the distributions and extents of the six regional patterns were characterized through their CADD features. The disease division index of every area fraction combination and the asymmetric index between the left and right lungs were also evaluated. A second SVM classifier was employed to classify the UIP and NSIP, and features were selected through sequential-forward floating feature selection. For the evaluation, 54 HRCT images of UIP (n = 26) and NSIP (n = 28) patients clinically diagnosed by a pulmonologist were included and evaluated. The classification accuracy was measured based on a fivefold cross-validation with 20 repetitions using random shuffling. For comparison, thoracic radiologists assessed each case using HRCT images without clinical information or diagnosis. The accuracies of the radiologists' decisions were 75 and 87%. The accuracies of the CADD system using different features ranged from 70 to 81%. Finally, the accuracy of the proposed CADD system after sequential-forward feature selection was 91%.

Selection of glucocorticoid-sensitive patients in interstitial lung disease secondary to connective tissue diseases population by radiomics

“…Many studies have reported that HRCT is very important in determining the extent and severity of ILD. 23 – 25 The distribution and manifestation of HRCT lesions may be related to the pathological type of ILD. 23 Patterns of ILD have been characterized by a few HRCT manifestations, including ground-glass opacities, interlobular septal thickening, and grid-like, honeycomb-like lesions.…”

“… 23 – 25 The distribution and manifestation of HRCT lesions may be related to the pathological type of ILD. 23 Patterns of ILD have been characterized by a few HRCT manifestations, including ground-glass opacities, interlobular septal thickening, and grid-like, honeycomb-like lesions. 13 , 26 However, a patient may present all types of lesions.…”

A lung graph model for the classification of interstitial lung diseases on CT images

“…Depeursinge et al 7 demonstrated the wealth of a regional approach for the computed diagnosis using a geometric atlas of the lungs and regional texture features. More recently, a study from Jun et al 8 highlighted the discriminative power of a texture comparison between upper-lower, anterior-posterior and central-peripheral parts of the lung. This paper builds on this previous work with a dataset of 450 labeled cases and with 3 classes of ILDs, Usual Interstitial Pneumonia (UIP), Non-Specific Interstitial Pneumonia (NSIP) and Chronic Hypersensitivity Pneumonitis (CHP).…”

Momentum

“…More recently, the development of therapeutic alternatives has brought about a new “moment” of growth, 9 moving us toward another point of change, in which the interpretation of imaging exams is no longer merely visual and tends to have a highly significant mathematical aspect in the aggregate. 10 It is still difficult to know what impact this will have on day-to-day clinical practice and patient management, given the existing therapeutic limitations, although it is not difficult to imagine that, with the development of new drugs, a more precise diagnostic analysis will allow individualized medicine to be used in the approach to patients with interstitial lung disease. The current issue of the JBP presents a pictorial essay that discusses, in a quite didactic manner, all of the certainties and inconsistencies that involve the diagnosis of usual interstitial pneumonia, the most prevalent form of idiopathic interstitial lung disease.…”

“… 9 Caminha-se para outro ponto de mudança, no qual a interpretação dos exames de imagem deixa de ser meramente visual e tende a ter um aspecto matemático agregado altamente significativo. 10 O impacto disso na clínica diária e no manejo dos pacientes é algo ainda difícil de especular, dadas as limitações terapêuticas existentes, mas não é difícil imaginar que, com o desenvolvimento de novos fármacos, a análise diagnóstica mais fina se aproximará da medicina personalizada na abordagem ao paciente com doença intersticial pulmonar. A presente edição do JBP traz um ensaio pictórico que discute, de forma bastante didática, todas as certezas e inconsistências que envolvem o diagnóstico da pneumonia intersticial usual, a forma mais prevalente das doenças intersticiais pulmonares idiopáticas.…”