Abstract
BackgroundThe differential diagnosis of tuberculous pleural effusion (TPE) is challenging. In recent years, artificial intelligence (AI) machine learning algorithms have started being used to an increasing extent in disease diagnosis due to the high level of efficiency, objectivity, and accuracy that they offer.MethodsData samples on 192 patients with TPE, 54 patients with parapneumonic pleural effusion (PPE), and 197 patients with malignant pleural effusion (MPE) were retrospectively collected. Based on 28 different features obtained via statistical analysis, TPE diagnostic models using four machine learning algorithms (MLAs), namely logistic regression, k-nearest neighbors (KNN), support vector machine (SVM) and random forest (RF) were established and their respective diagnostic performances were calculated. The respective diagnostic performances of each of the four algorithmic models were compared with that of pleural fluid adenosine deaminase (pfADA). Based on 12 features with the most significant impacts on the accuracy of the RF model, a new RF model was designed for clinical application. To demonstrate its external validity, a prospective study was conducted and the diagnostic performance of the RF model was calculated.ResultsThe respective sensitivity and specificity of each of the four TPE diagnostic models were as follows: logistic regression – 80.5 and 84.8%; KNN– 78.6 and 86.6%; SVM – 83.2 and 85.9%; and RF – 89.1 and 93.6%. The sensitivity and specificity of pfADA were 85.4 and 84.1%, respectively, at the best cut-off value of 17.5 U/L. RF was the superior method among the four MLAs, and was also superior to pfADA. The newly designed RF model (based on 12 out of 28 features) exhibited an acceptable performance rate for the diagnosis of TPE with a sensitivity and specificity of 90.6 and 92.3%, respectively. In the prospective study, its sensitivity and specificity were 100.0 and 90.0%, respectively.ConclusionsEstablishing a model for the diagnosis of TPE using RF resulted in a more effective, economical, and faster diagnostic method. This method could enable clinicians to diagnose and treat TPE more effectively.
Highlights
The differential diagnosis of tuberculous pleural effusion (TPE) is challenging
The following 28 features were introduced into the model: age, fever, cough, chest pain, anorexia, fatigue, night sweats, history of smoking, total blood WBC, neutrophil percentage (N%) in blood, lymphocyte percentage (L%) in blood, monocyte percentage (M%) in blood, platelet count (PLT), Erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), serum Lactate dehydrogenase (LDH), serum ADA, serum Carcinoembryonic antigen (CEA), bloody effusion, Rivalta test results, total WBC in pleural fluid, N% in pleural fluid, L% in pleural fluid, pleural fluid total protein, pleural fluid glucose concentration, pleural fluid LDH, pleural fluid adenosine deaminase (pfADA), pleural fluid CEA
Among the four algorithmic models, random forest (RF) is the superior method for diagnosing TPE, with a sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio (PLR), and accuracy higher than those of logistic regression, k-nearest neighbors (KNN), support vector machine (SVM), and pfADA
Summary
The differential diagnosis of tuberculous pleural effusion (TPE) is challenging. Artificial intelligence (AI) machine learning algorithms have started being used to an increasing extent in disease diagnosis due to the high level of efficiency, objectivity, and accuracy that they offer. Tuberculous pleurisy is a common disease that causes pleural effusion. Accurate diagnosis and timely treatment are vital. The gold standard in the diagnosis of tuberculous pleural effusion (TPE) derives from positive findings in pathogenic and pathological examinations. Neutrophils may predominate during the early stages of TPE [3], and the pfADA cut-off values for the diagnosis of TPE differ across several different studies [4, 5]. It is necessary to develop a method for the early diagnosis of TPE which is less invasive and more accurate
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
