J Melendez1, R H H M Philipsen1, P Chanda-Kapata2, V Sunkutu3, N Kapata2, B van Ginneken1. 1. Diagnostic Image Analysis Group, Radboud University Medical Center, Nijmegen, Thirona, Nijmegen, The Netherlands. 2. Ministry of Health Headquarters, Lusaka, Zambia; Centre for Tropical Medicine and Travel Medicine, Department of Infectious Diseases, Academic Medical Centre, University of Amsterdam, Amsterdam, The Netherlands. 3. University Teaching Hospital, Lusaka, Zambia.
Abstract
SETTING: Tuberculosis (TB) prevalence survey in Zambia between 2013 and 2014. OBJECTIVE: To compare the performance of automatic software (CAD4TB 5) in chest X-ray (CXR) reading with that of field (general practitioners) and central (radiologists) readers. DESIGN: A retrospective study comparing the performance of human and automatic reading was conducted. Two scenarios for central reading were evaluated: abnormalities not consistent with TB were considered to be 'normal' or 'abnormal'. Sputum culture was defined as the reference standard. Measures derived from receiver operating characteristic analysis were used to assess readers' performances. RESULTS: Of 46 099 participants, 23 838 cases included all survey information; of these, 106 cases were culture-confirmed TB-positive. The performance of CAD4TB 5 was similar to that of field and central readers. Although there were significant differences in specificity when compared with field readings (P = 0.002) and central readings considering the first scenario (P < 0.001), these differences were not substantial (93.2% vs. 92.6% and 98.4% vs. 99.6%, respectively).CONCLUSIONp: The performance of automatic CXR readings is comparable with that of human experts in a TB prevalence survey setting using culture as reference.
SETTING:Tuberculosis (TB) prevalence survey in Zambia between 2013 and 2014. OBJECTIVE: To compare the performance of automatic software (CAD4TB 5) in chest X-ray (CXR) reading with that of field (general practitioners) and central (radiologists) readers. DESIGN: A retrospective study comparing the performance of human and automatic reading was conducted. Two scenarios for central reading were evaluated: abnormalities not consistent with TB were considered to be 'normal' or 'abnormal'. Sputum culture was defined as the reference standard. Measures derived from receiver operating characteristic analysis were used to assess readers' performances. RESULTS: Of 46 099 participants, 23 838 cases included all survey information; of these, 106 cases were culture-confirmed TB-positive. The performance of CAD4TB 5 was similar to that of field and central readers. Although there were significant differences in specificity when compared with field readings (P = 0.002) and central readings considering the first scenario (P < 0.001), these differences were not substantial (93.2% vs. 92.6% and 98.4% vs. 99.6%, respectively).CONCLUSIONp: The performance of automatic CXR readings is comparable with that of human experts in a TB prevalence survey setting using culture as reference.
Authors: J Melendez; L Hogeweg; C I Sánchez; R H H M Philipsen; R W Aldridge; A C Hayward; I Abubakar; B van Ginneken; A Story Journal: Int J Tuberc Lung Dis Date: 2018-05-01 Impact factor: 2.373
Authors: Miriam Harris; Amy Qi; Luke Jeagal; Nazi Torabi; Dick Menzies; Alexei Korobitsyn; Madhukar Pai; Ruvandhi R Nathavitharana; Faiz Ahmad Khan Journal: PLoS One Date: 2019-09-03 Impact factor: 3.240
Authors: Jana Fehr; Stefan Konigorski; Stephen Olivier; Resign Gunda; Ashmika Surujdeen; Dickman Gareta; Theresa Smit; Kathy Baisley; Sashen Moodley; Yumna Moosa; Willem Hanekom; Olivier Koole; Thumbi Ndung'u; Deenan Pillay; Alison D Grant; Mark J Siedner; Christoph Lippert; Emily B Wong Journal: NPJ Digit Med Date: 2021-07-02
Authors: Keelin Murphy; Shifa Salman Habib; Syed Mohammad Asad Zaidi; Saira Khowaja; Aamir Khan; Jaime Melendez; Ernst T Scholten; Farhan Amad; Steven Schalekamp; Maurits Verhagen; Rick H H M Philipsen; Annet Meijers; Bram van Ginneken Journal: Sci Rep Date: 2020-03-26 Impact factor: 4.996