BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Simulated X-ray Radiography for Synthetic TB Data Generation
DTSTART;VALUE=DATE-TIME:20241121T124500Z
DTEND;VALUE=DATE-TIME:20241121T130000Z
DTSTAMP;VALUE=DATE-TIME:20261010T203623Z
UID:indico-contribution-9823@events.saip.org.za
DESCRIPTION:Speakers: Preveshin Maduray (University of Johannesburg)\n1. I
 ntroduction \n\nThe project focuses on training Machine Learning (ML) algo
 rithms to detect pulmonary Tuberculosis (TB) infections in clinical imagin
 g. We have identified 5 publicly available datasets of TB infected lung ra
 diographs in medical literature. These datasets have been used repeatedly 
 to train ML algorithms to detect TB.1 The main reason for improvements in 
 accuracy of TB detection has been the use of increasingly sophisticated al
 gorithms.1234 To introduce a high volume of new data with known TB indicat
 ors\, we propose to generate synthetic data from high-resolution digital t
 win lungs. The study will leverage Hierarchical Phase-Contrast Tomography 
 (HiP-CT) scans which are produced at the European Synchrotron Research Fac
 ility (ESRF) beamline BM18. These are high-resolution 3D scans (~20 µm) o
 f whole organs\, with regions of sub 2.5 µm resolution. Scans will be tak
 en of ex-vivo healthy and TB-infected lungs. The data will be processed in
 to a large number of 3D models that include a known set of TB indicators. 
 These models will be reprojected into simulated Chest X-rays (CXR) with so
 ftware using the Geant4 toolkit which simulates the passage of particles t
 hrough matter. These simulated radiographs with their labelled pathologies
  will then be used to train classification algorithms for the detection of
  TB in the lung. This method could be extended to various other organs in 
 future\, and as such can assist us in improving the diversity of datasets 
 utilized by the ML community.\n\n2. Results\n\nFigure 1 is a CXR of a pati
 ent infected with TB. The dark region on the left side is a pulmonary cavi
 tation. This occurs when normal lung tissue dies from the infection\, thus
  becoming gas filled. Figure 2 is a demonstration of simulating a CXR usin
 g a low-resolution 2D model. The X-rays passing from source to detector in
 teract with several tissue classes and scatter realistically. The detector
  element can record the X-ray dose absorbed to build a 2D radiograph like 
 in figure 1. We intend to use high resolution HiP-CT data which\, due to t
 he non-linear progression of TB\, can provide a detailed understanding of 
 the evolution of pathologies down to a cellular level. We aim to combine s
 cans into many digital organs which will exhibit differing presentations o
 f the infection and include truth data allowing optimised training of ML a
 lgorithms.\n\n                                    \n\n\n\n3. References\n\
 n[1]	T. Rahman et al.\, "Reliable Tuberculosis Detection Using Chest X-Ray
  With Deep Learning\, Segmentation and Visualization\," 2020\, doi: 10.110
 9/ACCESS.2020.3031384\n[2]	Showkatian E\, Salehi M\, Ghaffari H\, Reiazi R
 \, Sadighi N. Deep learning-based automatic detection of tuberculosis dise
 ase in chest X-ray images. Pol J Radiol. 2022\, doi: 10.5114/pjr.2022.1134
 35. PMID: 35280947\; PMCID: PMC8906182\n[3]	Rajesh\, T. Babu\, R. R. Nair 
 and P. Pechetti\, "Detection of Tuberculosis using a Multi-model Classific
 ation Approach on CXR Images\," 2022\, doi: 10.1109/AIDE57180.2022.1006067
 5\n[4]	D. Capellán-Martín\, J. J. Gómez-Valverde\, D. Bermejo-Peláez a
 nd M. J. Ledesma-Carbayo\, "A Lightweight\, Rapid and Efficient Deep Convo
 lutional Network for Chest X-Ray Tuberculosis Detection\," 2023\, doi: 10.
 1109/ISBI53787.2023.10230500\n[5]	Ravimohan\, S.\, Kornfeld\, H.\, Weissma
 n\, D.\, & Bisson\, G. P. (2018). Tuberculosis and lung damage: from epide
 miology to pathophysiology. https://doi.org/10.1183/16000617.0077-2017\n\n
 https://events.saip.org.za/event/249/contributions/9823/
LOCATION:
URL:https://events.saip.org.za/event/249/contributions/9823/
END:VEVENT
END:VCALENDAR