Gerrand, Jonathan DWilliams, Quentin RLunga, DPantanowitz, AMadhi, SMahomed, N2017-09-222017-09-222017-07Gerrand, J.D., Williams, Q.R., Lunga, D. et al. 2017. Paediatric frontal chest radiograph screening with fine-tuned convolutional neural networks. Annual Conference on Medical Image Understanding and Analysis (MIUA), 11-13 July 2017, Edinburgh, United Kingdom, pp. 850-861978-3-319-60964-5978-3-319-60963-81865-0937https://link.springer.com/chapter/10.1007/978-3-319-60964-5_74doi.org/10.1007/978-3-319-60964-5_74http://hdl.handle.net/10204/9594Copyright: 2017 Springer International. Due to copyright restrictions, the attached PDF file only contains the abstract of the full text item. For access to the full text item, kindly consult the publisher's website.Within developing countries, there is a realistic need for technologies that can assist medical practitioners in meeting the increasing demand for patient screening and monitoring. To this end, computer aided diagnosis (CAD) based approaches to chest radiograph screening can be utilised in areas where there is a high burden of diseases such as tuberculosis and pneumonia. In this work, we investigate the efficacy of a purely data-driven approach to chest radiograph classification through the use of fine-tuned convolutional neural networks (CNN). We use two popular CNN models that are pre-trained on a large natural image dataset and two distinct datasets containing paediatric and adult radiographs respectively. Evaluation is performed using a 5-fold cross-validation analysis at an image level. The promising results, with top AUC metrics of 0.87 and 0.84 for the respective datasets, along with several characteristics of our data-driven approach motivate for the use of fine-tuned CNN models within this application of CAD.enComputer aided diagnosisConvolutional neural networksChest radiograph screeningFine-tuningConference PresentationGerrand, J. D., Williams, Q. R., Lunga, D., Pantanowitz, A., Madhi, S., & Mahomed, N. (2017). Paediatric frontal chest radiograph screening with fine-tuned convolutional neural networks. Springer. http://hdl.handle.net/10204/9594Gerrand, Jonathan D, Quentin R Williams, D Lunga, A Pantanowitz, S Madhi, and N Mahomed. "Paediatric frontal chest radiograph screening with fine-tuned convolutional neural networks." (2017): http://hdl.handle.net/10204/9594Gerrand JD, Williams QR, Lunga D, Pantanowitz A, Madhi S, Mahomed N, Paediatric frontal chest radiograph screening with fine-tuned convolutional neural networks; Springer; 2017. http://hdl.handle.net/10204/9594 .TY - Conference Presentation AU - Gerrand, Jonathan D AU - Williams, Quentin R AU - Lunga, D AU - Pantanowitz, A AU - Madhi, S AU - Mahomed, N AB - Within developing countries, there is a realistic need for technologies that can assist medical practitioners in meeting the increasing demand for patient screening and monitoring. To this end, computer aided diagnosis (CAD) based approaches to chest radiograph screening can be utilised in areas where there is a high burden of diseases such as tuberculosis and pneumonia. In this work, we investigate the efficacy of a purely data-driven approach to chest radiograph classification through the use of fine-tuned convolutional neural networks (CNN). We use two popular CNN models that are pre-trained on a large natural image dataset and two distinct datasets containing paediatric and adult radiographs respectively. Evaluation is performed using a 5-fold cross-validation analysis at an image level. The promising results, with top AUC metrics of 0.87 and 0.84 for the respective datasets, along with several characteristics of our data-driven approach motivate for the use of fine-tuned CNN models within this application of CAD. DA - 2017-07 DB - ResearchSpace DP - CSIR KW - Computer aided diagnosis KW - Convolutional neural networks KW - Chest radiograph screening KW - Fine-tuning LK - https://researchspace.csir.co.za PY - 2017 SM - 978-3-319-60964-5 SM - 978-3-319-60963-8 SM - 1865-0937 T1 - Paediatric frontal chest radiograph screening with fine-tuned convolutional neural networks TI - Paediatric frontal chest radiograph screening with fine-tuned convolutional neural networks UR - http://hdl.handle.net/10204/9594 ER -