Resource Management through Artificial Intelligence in Screening Programs—Key for the Successful Elimination of Hepatitis C

Abstract

Background: The elimination of the Hepatitis C virus (HCV) will only be possible if rapid and efficient actions are taken. Artificial neural networks (ANNs) are computing systems based on the topology of the biological brain, containing connected artificial neurons that can be tasked with solving medical problems. Aim: We expanded the previously presented HCV micro-elimination project started in September 2020 that aimed to identify HCV infection through coordinated screening in asymptomatic populations and developed two ANN models able to identify at-risk subjects selected through a targeted questionnaire. Material and method: Our study included 14,042 screened participants from a southwestern region of Oltenia, Romania. Each participant completed a 12-item questionnaire along with anti-HCV antibody rapid testing. Hepatitis-C-positive subjects were linked to care and ultimately could receive antiviral treatment if they had detectable viremia. We built two ANNs, trained and tested on the dataset derived from the questionnaires and then used to identify patients in a similar, already existing dataset. Results: We found 114 HCV-positive patients (81 females), resulting in an overall prevalence of 0.81%. We identified sharing personal hygiene items, receiving blood transfusions, having dental work or surgery and re-using hypodermic needles as significant risk factors. When used on an existing dataset of 15,140 persons (119 HCV cases), the first ANN models correctly identified 97 (81.51%) HCV-positive subjects through 13,401 tests, while the second ANN model identified 81 (68.06%) patients through only 5192 tests. Conclusions: The use of ANNs in selecting screening candidates may improve resource allocation and prioritize cases more prone to severe disease.

Opportunities and challenges for the application of artificial intelligence paradigms into the management of endemic viral infections: The example of Chronic Hepatitis C Virus

“…The AI/ML applications introduced in this review showed comparable or better performance to the traditional third‐generation rapid HCV‐Ab tests 64,91 . The only outlier was the ML model by Butaru 40 , which had a very low predictive power for detecting positive HCV infections. However, it performed similarly to rapid HCV‐Ab as a Point‐of‐Care (POC) tool to exclude HCV infection (Table 2).…”

“…Wei, L. & Ying, X., 2011 34 ; Wang, M. et al, 2012 35 ; Wei, Y. et al, 2016 36 ; Malik, A. A. et al, 2021 37 ; Wei, Y. et al, 2019 38 ; Reiser, M. et al, 2019 39 ; Butaru, A. E. et al, 2022 40 ; Villarreal, Y. R. et al, 2019 41 ; Doyle O. M. et al, 2020 42 ; Edeh, M. O. et al, 2022 43 ; Chirikov, V. V. et al, 2018 44 ; Park, H. et al, 2022 45 ; Janczewska, E. et al, 2021 46 ; Haga, H. et al, 2020 47 ; Churkin, A. et al, 2022 48 ; Rivero‐Juárez, A. et al, 2020 49 ; Oselio, B. et al, 2022 50 ; Weber, T. et al, 2022 51 ; Itami‐Matsumoto et al, 2019 52 ; Hashem, S. et al, 2020 53 ; Audureau, E. et al, 2020 54 ; G. Wong et al, 2022 55 ; Hershberger, C. E. et al, 2021 56 …”

“…The preclinical fields involved are the development of novel HCV treatments and HCV/human immunodeficiency virus-1 (HIV-1) coinhibitors, [34][35][36][37][38] while the clinical aspects covered are the early detection of HCV infection, [39][40][41][42][43] the prediction of DAAs treatment outcomes. [44][45][46][47][48][49][50] The development of HCV vaccines, 51 and the early diagnosis of HCV-related HCC.…”

“…Using a 12‐item questionnaire, Butaru 40 built a deep artificial neural network (DANN) that could predict HCV infection 62 . The DANN predictor revealed a comparable specificity and negative predictive value (NPV) to those reported by the third‐generation rapid HCV‐antibody tests 63 (99.7% vs. 99.7% and 99.2% vs. 99.99%, respectively, promoting it as a rapid non‐costly tool to exclude HCV infection in high‐prevalence settings with limited resources, 64 Another questionnaire was employed by Villarreal 41 to train an RF classifier that can identify HCV infections among drug abusers, one of the HCV key populations, which constitutes 23%–39% of new HCV infections and around one‐third of HCV‐related deaths globally 65,66 .…”

An Automated Method for Classifying Liver Lesions in Contrast-Enhanced Ultrasound Imaging Based on Deep Learning Algorithms

“…Therefore, random samples from each class in the dataset were duplicated. This technique has been proven to be effective when dealing with class imbalance [ 14 , 15 ]. The dataset was split in 70% samples for training and 30% samples for validation.…”