AI-Based Analysis of Ziehl–Neelsen-Stained Sputum Smears for <i>Mycobacterium tuberculosis</i> as a Screening Method for Active Tuberculosis
Arief Budi Witarto,
Bogdan Ceachi,
Cristiana Popp,
Sabina Zurac,
Ioana Cristina Daha,
Flora Eka Sari,
Nirawan Putranto,
Satria Pratama,
Benyamin P. Octavianus,
Luciana Nichita,
Julian Gerald Dcruz,
Cristian Mogodici,
Mirela Cioplea,
Liana Sticlaru,
Mihai Busca,
Oana Stefan,
Irina Tudor,
Carmen Dumitru,
Alexandra Vilaia,
Alexandra Bastian,
Gheorghita Jugulete,
Gyula Laszlo Fekete,
Petronel Mustatea
Affiliations
Arief Budi Witarto
Medical Biodefense Research Center, Faculty of Medicine and Health Sciences, Indonesia Defense University, Kawasan IPSC Sentul, Sukahati, Kecamatan Citeureup, Kabupaten Bogor 16810, Indonesia
Bogdan Ceachi
Department of Pathology, Colentina University Hospital, 21 Stefan Cel Mare Str., Sector 2, 020125 Bucharest, Romania
Cristiana Popp
Department of Pathology, Colentina University Hospital, 21 Stefan Cel Mare Str., Sector 2, 020125 Bucharest, Romania
Sabina Zurac
Department of Pathology, Colentina University Hospital, 21 Stefan Cel Mare Str., Sector 2, 020125 Bucharest, Romania
Ioana Cristina Daha
Department of Cardiology, Colentina University Hospital, 21 Stefan Cel Mare Str., Sector 2, 020125 Bucharest, Romania
Flora Eka Sari
Medical Biodefense Research Center, Faculty of Medicine and Health Sciences, Indonesia Defense University, Kawasan IPSC Sentul, Sukahati, Kecamatan Citeureup, Kabupaten Bogor 16810, Indonesia
Nirawan Putranto
Medical Biodefense Research Center, Faculty of Medicine and Health Sciences, Indonesia Defense University, Kawasan IPSC Sentul, Sukahati, Kecamatan Citeureup, Kabupaten Bogor 16810, Indonesia
Satria Pratama
Medical Biodefense Research Center, Faculty of Medicine and Health Sciences, Indonesia Defense University, Kawasan IPSC Sentul, Sukahati, Kecamatan Citeureup, Kabupaten Bogor 16810, Indonesia
Benyamin P. Octavianus
Badan Kesira Gerindra, Jl. Harsono RM No. 54, Ragunan, Pasar Minggu, Jakarta Selatan 12160, Indonesia
Luciana Nichita
Department of Pathology, Colentina University Hospital, 21 Stefan Cel Mare Str., Sector 2, 020125 Bucharest, Romania
Julian Gerald Dcruz
Zaya Artificial Intelligence SRL, 9A Stefan Cel Mare Str., 077190 Voluntari, Romania
Cristian Mogodici
Department of Pathology, Colentina University Hospital, 21 Stefan Cel Mare Str., Sector 2, 020125 Bucharest, Romania
Mirela Cioplea
Department of Pathology, Colentina University Hospital, 21 Stefan Cel Mare Str., Sector 2, 020125 Bucharest, Romania
Liana Sticlaru
Department of Pathology, Colentina University Hospital, 21 Stefan Cel Mare Str., Sector 2, 020125 Bucharest, Romania
Mihai Busca
Department of Pathology, Colentina University Hospital, 21 Stefan Cel Mare Str., Sector 2, 020125 Bucharest, Romania
Oana Stefan
Department of Pathology, Colentina University Hospital, 21 Stefan Cel Mare Str., Sector 2, 020125 Bucharest, Romania
Irina Tudor
Department of Pathology, Colentina University Hospital, 21 Stefan Cel Mare Str., Sector 2, 020125 Bucharest, Romania
Carmen Dumitru
Department of Pathology, Colentina University Hospital, 21 Stefan Cel Mare Str., Sector 2, 020125 Bucharest, Romania
Alexandra Vilaia
Department of Pathology, University of Medicine and Pharmacy Carol Davila, 37 Dionisie Lupu Str., Sector 1, 020021 Bucharest, Romania
Alexandra Bastian
Department of Pathology, Colentina University Hospital, 21 Stefan Cel Mare Str., Sector 2, 020125 Bucharest, Romania
Gheorghita Jugulete
Department of Infectious Diseases, National Institute for Infectious Diseases “Prof. Dr. Matei Balș”, 1 Dr. Calistrat Grozovici Str., Sector 2, 021105 Bucharest, Romania
Gyula Laszlo Fekete
Department of Dermatology, George Emil Palade University of Medicine, Pharmacy, Science and Technology of Targu Mures, 38 Gh. Marinescu Str., 540142 Targu Mures, Romania
Petronel Mustatea
Department of Pathology, Colentina University Hospital, 21 Stefan Cel Mare Str., Sector 2, 020125 Bucharest, Romania
Tuberculosis is the primary cause of death due to infection in the world. Identification of Mycobacterium tuberculosis in sputum is a diagnostic test, which can be used in screening programs—especially in countries with a high incidence of tuberculosis—to identify and treat those persons with the highest risk of disseminating the infection. We previously developed an algorithm which is able to automatically detect mycobacteria on tissue; in particular, our algorithm identified acid-fast bacilli on tissue with 100% specificity, 95.65% sensitivity, and 98.33% accuracy. We tested this algorithm on 1059 Ziehl–Neelsen-stained sputum smears to evaluate its results as a possible tool for screening. The results were displayed as a heat map of 32 × 32 pixel patches. Analysis of the positive patches revealed a good specificity (86.84%) and 100% sensitivity for patches with a level of confidence over 90; furthermore, the accuracy remained over 95% for all levels of confidence over 80, except the class (95–100]. The modest specificity is caused by the peculiarities of smears (uneven thickness, dust contamination, lack of coverslip). We will train the algorithm on sputum smears to increase the specificity to over 95%. However, as our algorithm showed no false negatives, it is suitable for screening.
