Computer aided detection of tuberculosis using two classifiers

Abstract

Abstracts Objectives Tuberculosis caused by Mycobacterium tuberculosis have been a major challenge for medical and healthcare sectors in many underdeveloped countries with limited diagnosis tools. Tuberculosis can be detected from microscopic slides and chest X-ray but as a result of the high cases of tuberculosis, this method can be tedious for both microbiologist and Radiologist and can lead to miss-diagnosis. The main objective of this study is to addressed these challenges by employing Computer Aided Detection (CAD) using Artificial Intelligence-driven models which learn features based on convolution and result in an output with high accuracy. Method In this paper, we described automated discrimination of X-ray and microscopic slide images of tuberculosis into positive and negative cases using pretrained AlexNet Models. The study employed Chest X-ray dataset made available on Kaggle repository and microscopic slide images from both Near East university hospital and Kaggle repository. Results For classification of tuberculosis and healthy microscopic slide using AlexNet+Softmax, the model achieved accuracy of 98.14%. For classification of tuberculosis and healthy microscopic slide using AlexNet+SVM, the model achieved 98.73% accuracy. For classification of tuberculosis and healthy chest X-ray images using AlexNet+Softmax, the model achieved accuracy of 98.19%. For classification of tuberculosis and healthy chest X-ray images using AlexNet+SVM, the model achieved 98.38% accuracy. Conclusion The result obtained has shown to outperformed several studies in the current literature. Future studies will attempt to integrate Internet of Medical Things (IoMT) for the design of IoMT/AI-enabled platform for detection of Tuberculosis from both X-ray and Microscopic slide images.

Large‐scaled detection of COVID‐19 from X‐ray using transfer learning

“…18 The application of CAD systems in medical imaging is assisting medical professionals conduct more reliable, accurate, fast, and cost-effective diagnoses. Over the last two decades, scientists have applied CAD technology for the detection of diseases, such as non-COVID-19 pneumonia, COVID-19 pneumonia, 6 tuberculosis, 19 ad cancer (breast, prostate, melanoma etc. ).…”

Current Technologies for Detection of COVID-19: Biosensors, Artificial Intelligence and Internet of Medical Things (IoMT): Review

“…In order to address these issues, scientists merge radiographic imaging with computer applications to developed computer aided diagnosis which allow screening of thousand images with high accuracy, precision and specificity [8][9]. CAD has shown to aid medical experts in the past for the detection of different types of cancer such as breast cancer [10], colon cancer [11], prostate cancer [12], tuberculosis [13], bacterial pneumonia [14], non-COVID-19 viral pneumonia [15], skin disease [16].…”

Current Technologies for Detection of COVID-19: Biosensors, Artificial Intelligence and Internet of Medical Things (IoMT): Review

“…In order to address these issues, scientists merge radiographic imaging with computer applications to develop computer-aided diagnosis (CAD), which allows screening of thousands of images with high accuracy, precision, and specificity [ 10 , 11 ]. CAD has been shown to aid medical experts in the past in the detection of different types of cancer, such as breast cancer [ 12 ], colon cancer [ 13 ], prostate cancer [ 14 ], brain cancer [ 15 ], tuberculosis [ 16 ], bacterial pneumonia [ 17 ], non-COVID-19 viral pneumonia [ 18 ], and skin diseases [ 19 ].…”