Machine learning prediction models for clinical management of blood-borne viral infections: a systematic review of current applications and future impact.

BACKGROUND: Machine learning (ML) prediction models to support clinical management of blood-borne viral infections are becoming increasingly abundant in medical literature, with a number of competing models being developed for the same outcome or target population. However, evidence on the quality of these ML prediction models are limited. OBJECTIVE: This study aimed to evaluate the development and quality of reporting of ML prediction models that could facilitate timely clinical management of blood-borne viral infections. METHODS: We conducted narrative evidence synthesis following the synthesis without meta-analysis guidelines. We searched PubMed and Cochrane Central Register of Controlled Trials for all studies applying ML models for predicting clinical outcomes associated with hepatitis B virus (HBV), human immunodeficiency virus (HIV), or hepatitis C virus (HCV). RESULTS: We found 33 unique ML prediction models aiming to support clinical decision making. Overall, 12 (36.4%) focused on HBV, 10 (30.3%) on HCV, 10 on HIV (30.3%) and two (6.1%) on co-infection. Among these, six (18.2%) addressed the diagnosis of infection, 16 (48.5%) the prognosis of infection, eight (24.2%) the prediction of treatment response, two (6.1%) progression through a cascade of care, and one (3.03%) focused on the choice of antiretroviral therapy (ART). Nineteen prediction models (57.6%) were developed using data from high-income countries. Evaluation of prediction models was limited to measures of performance. Detailed information on software code accessibility was often missing. Independent validation on new datasets and/or in other institutions was rarely done. CONCLUSION: Promising approaches for ML prediction models in blood-borne viral infections were identified, but the lack of robust validation, interpretability/explainability, and poor quality of reporting hampered their clinical relevance. Our findings highlight important considerations that can inform standard reporting guidelines for ML prediction models in the future (e.g., TRIPOD-AI), and provides critical data to inform robust evaluation of the models.

Routine blood test markers for predicting liver disease post HBV infection: precision pathology and pattern recognition.

BACKGROUND: Early stages of hepatitis B virus (HBV) infection usually involve inflammation of the liver. Patients with chronic infection have an increased risk of progressive liver fibrosis, cirrhosis, and life-threatening clinical complications of end-stage hepatocellular carcinoma (HCC). CONTENT: Early diagnosis of hepatic fibrosis and timely clinical management are critical to controlling disease progression and decreasing the burden of end-stage liver cancer. Fibrosis staging, through its current gold standard, liver biopsy, improves patient outcomes, but the clinical procedure is invasive with unpleasant post-procedural complications. Routine blood test markers offer promising diagnostic potential for early detection of liver disease without biopsy. There is a plethora of candidate routine blood test markers that have gone through phases of biomarker validation and have shown great promise, but their current limitations include a predictive ability that is limited to only a few stages of fibrosis. However, the advent of machine learning, notably pattern recognition, presents an opportunity to refine blood-based non-invasive models of hepatic fibrosis in the future. SUMMARY: In this review, we highlight the current landscape of routine blood-based non-invasive models of hepatic fibrosis, and appraise the potential application of machine learning (pattern recognition) algorithms to refining these models and optimising clinical predictions of HBV-associated liver disease. OUTLOOK: Machine learning via pattern recognition algorithms takes data analytics to a new realm, and offers the opportunity for enhanced multi-marker fibrosis stage prediction using pathology profile that leverages information across patient routine blood tests.

Clinical Validity of a Machine Learning Decision Support System for Early Detection of Hepatitis B Virus: A Binational External Validation Study.

HepB LiveTest is a machine learning decision support system developed for the early detection of hepatitis B virus (HBV). However, there is a lack of evidence on its generalisability. In this study, we aimed to externally assess the clinical validity and portability of HepB LiveTest in predicting HBV infection among independent patient cohorts from Nigeria and Australia. The performance of HepB LiveTest was evaluated by constructing receiver operating characteristic curves and estimating the area under the curve. Delong's method was used to estimate the 95% confidence interval (CI) of the area under the receiver-operating characteristic curve (AUROC). Compared to the Australian cohort, patients in the derivation cohort of HepB LiveTest and the hospital-based Nigerian cohort were younger (mean age, 45.5 years vs. 38.8 years vs. 40.8 years, respectively; p < 0.001) and had a higher incidence of HBV infection (1.9% vs. 69.4% vs. 57.3%). In the hospital-based Nigerian cohort, HepB LiveTest performed optimally with an AUROC of 0.94 (95% CI, 0.91-0.97). The model provided tailored predictions that ensured most cases of HBV infection did not go undetected. However, its discriminatory measure dropped to 0.60 (95% CI, 0.56-0.64) in the Australian cohort. These findings indicate that HepB LiveTest exhibits adequate cross-site transportability and clinical validity in the hospital-based Nigerian patient cohort but shows limited performance in the Australian cohort. Whilst HepB LiveTest holds promise for reducing HBV prevalence in underserved populations, caution is warranted when implementing the model in older populations, particularly in regions with low incidence of HBV infection.

The development of a machine learning algorithm for early detection of viral hepatitis B infection in Nigerian patients.

Access to Hepatitis B Virus (HBV) testing for people in low-resource settings has long been challenging due to the gold standard, enzyme immunoassay, being prohibitively expensive, and requiring specialised skills and facilities that are not readily available, particularly in remote and isolated laboratories. Routine pathology data in tandem with cutting-edge machine learning shows promising diagnostic potential. In this study, recursive partitioning ("trees") and Support Vector Machines (SVMs) were applied to interrogate patient dataset (n = 916) that comprised results for Hepatitis B Surface Antigen (HBsAg) and routine clinical chemistry and haematology blood tests. These algorithms were used to develop a predictive diagnostic model of HBV infection. Our SVM-based diagnostic model of infection (accuracy = 85.4%, sensitivity = 91%, specificity = 72.6%, precision = 88.2%, F1-score = 0.89, Area Under the Receiver Operating Curve, AUC = 0.90) proved to be highly accurate for discriminating HBsAg positive from negative patients, and thus rivals with immunoassay. Therefore, we propose a predictive model based on routine blood tests as a novel diagnostic for early detection of HBV infection. Early prediction of HBV infection via routine pathology markers and pattern recognition algorithms will offer decision-support to clinicians and enhance early diagnosis, which is critical for optimal clinical management and improved patient outcomes.

Hepatitis B virus infection in Nigeria: a systematic review and meta-analysis of data published between 2010 and 2019.

BACKGROUND: Hepatitis B virus (HBV) is an infectious disease of global significance, causing a significant health burden in Africa due to complications associated with infection, such as cirrhosis and liver cancer. In Nigeria, which is considered a high prevalence country, estimates of HBV cases are inconsistent, and therefore additional clarity is required to manage HBV-associated public health challenges. METHODS: A systematic review of the literature (via PubMed, Advanced Google Scholar, African Index Medicus) was conducted to retrieve primary studies published between 1 January 2010 and 31 December 2019, with a random-effects model based on proportions used to estimate the population-based prevalence of HBV in the Nigerian population. RESULTS: The final analyses included 47 studies with 21,702 participants that revealed a pooled prevalence of 9.5%. A prevalence estimate above 8% in a population is classified as high. Sub-group analyses revealed the highest HBV prevalence in rural settings (10.7%). The North West region had the highest prevalence (12.1%) among Nigeria's six geopolitical zones/regions. The estimate of total variation between studies indicated substantial heterogeneity. These variations could be explained by setting and geographical region. The statistical test for Egger's regression showed no evidence of publication bias (p = 0.879). CONCLUSIONS: We present an up-to-date review on the prevalence of HBV in Nigeria, which will provide critical data to optimise and assess the impact of current prevention and control strategies, including disease surveillance and diagnoses, vaccination policies and management for those infected.

One Health Approach: A Data-Driven Priority for Mitigating Outbreaks of Emerging and Re-Emerging Zoonotic Infectious Diseases.

This paper discusses the contributions that One Health principles can make in improving global response to zoonotic infectious disease. We highlight some key benefits of taking a One Health approach to a range of complex infectious disease problems that have defied a more traditional sectoral approach, as well as public health policy and practice, where gaps in surveillance systems need to be addressed. The historical examples demonstrate the scope of One Health, partly from an Australian perspective, but also with an international flavour, and illustrate innovative approaches and outcomes with the types of collaborative partnerships that are required.

Atypical clinical presentation of Ebola virus disease in pregnancy: Implications for clinical and public health management.

BACKGROUND: Between December 2013 and June 2016, West Africa experienced the largest Ebola virus disease (EVD) outbreak in history. Understanding EVD in pregnancy is important for EVD clinical screening and infection prevention and control. METHODS: We conducted a review of medical records and EVD investigation reports from three districts in Sierra Leone. We report the clinical presentations and maternal and fetal outcomes of six pregnant women with atypical EVD, and subsequent transmission events from perinatal care. RESULTS: The six women (ages 18-38) were all in the third trimester. Each presented with signs and symptoms initially attributed to pregnancy. None met EVD case definition; only one was known at presentation to be a contact of an EVD case. Five women died, and all six fetuses/neonates died. These cases resulted in at least 35 additional EVD cases. CONCLUSIONS: These cases add to the sparse literature focusing on pregnant women with EVD, highlighting challenges and implications for outbreak control. Infected newborns may also present atypically and may shed virus while apparently asymptomatic. Pregnant women identified a priori as contacts of EVD cases require special attention and planning for obstetrical care.

Evaluation of the early warning, alert and response system after Cyclone Winston, Fiji, 2016.

OBJECTIVE: To assess the performance of an early warning, alert and response system (EWARS) developed by the World Health Organization (WHO) - EWARS in a Box - that was used to detect and control disease outbreaks after Cyclone Winston caused destruction in Fiji on 20 February 2016. METHODS: Immediately after the cyclone, Fiji's Ministry of Health and Medical Services, supported by WHO, started to implement EWARS in a Box, which is a smartphone-based, automated, early warning surveillance system for rapid deployment during health emergencies. Both indicator-based and event-based surveillance were employed. The performance of the system between 7 March and 29 May 2016 was evaluated. Users' experience with the system was assessed in interviews using a semi-structured questionnaire and by a cross-sectional survey. The system's performance was assessed using data from the EWARS database. FINDINGS: Indicator-based surveillance recorded 34 113 cases of the nine syndromes under surveillance among 326 861 consultations. Three confirmed outbreaks were detected, and no large outbreak was missed. Users were satisfied with the performance of EWARS and judged it useful for timely monitoring of disease trends and outbreak detection. The system was simple, stable and flexible and could be rapidly deployed during a health emergency. The automated collation, analysis and dissemination of data reduced the burden on surveillance teams, saved human resources, minimized human error and ensured teams could focus on public health responses. CONCLUSION: In Fiji, EWARS in a Box was effective in strengthening disease surveillance during a national emergency and was well regarded by users.

Vancomycin-resistant Enterococcus (VRE) outbreak in a neonatal intensive care unit and special care nursery at a tertiary-care hospital in Australia-A retrospective case-control study.

OBJECTIVE: We investigated the risk factors and origins of the first known occurrence of VRE colonization in the neonatal intensive care unit (NICU) at the Canberra Hospital. DESIGN: A retrospective case-control study. SETTING: A 21-bed neonatal intensive care unit (NICU) and a 15-bed special care nursey (SCN) in a tertiary-care adult and pediatric hospital in Australia. PATIENTS: All patients admitted to the NICU and SCN over the outbreak period: January-May 2017. Of these, 14 were colonized with vancomycin-resistant Enterococcus (VRE) and 77 were noncolonized. METHODS: Demographic and clinical variables of cases and controls were compared to evaluate potential risk factors for VRE colonization. Whole-genome sequencing of the VRE isolates was used to determine the origin of the outbreak strain. RESULTS: Swift implementation of wide-ranging infection control measures brought the outbreak under control. Multivariate logistic regression revealed a strong association between early gestational age and VRE colonization (odds ratio [OR], 3.68; 95% confidence interval [CI], 1.94-7.00). Whole-genome sequencing showed the isolates to be highly clonal Enterococcus faecium ST1421 harboring a vanA gene and to be closely related to other ST1421 previously sequenced from the Canberra Hospital and the Australian Capital Territory. CONCLUSION: The colonization of NICU patients was with a highly successful clone endemic to the Canberra Hospital likely introduced into the NICU environment from other wards, with subsequent cross-contamination spreading among the neonate patients. Use of routine surveillance screening may have identified colonization at an earlier stage and have now been implemented on a 6-monthly schedule.

An outbreak investigation of paediatric severe acute respiratory infections requiring admission to intensive care units - Fiji, May 2016.

INTRODUCTION: Influenza-associated severe acute respiratory infections (SARI) are a major contributor to global morbidity and mortality. In response to a cluster of SARI cases and deaths in pregnant women, with two deceased cases testing positive for influenza A(H1N1)pdm09, an investigation was initiated to determine whether there was an increase of paediatric SARI cases admitted to divisional hospital intensive care units in Fiji in may 2016 compared to May 2013-2015. METHODS: Retrospective case finding was conducted at the paediatric intensive care units (PICUs) in Fiji's three divisional hospitals. Data were collected from 1 January 2013 to 26 May 2016. Cases were identified using a list of clinical diagnoses compatible with SARI. RESULTS: A total of 632 cases of paediatric SARI with complete details were identified. The median age of cases was 6 months (Interquartile range: 2-14 months). Children aged less than 5 years had a higher rate of paediatric SARI requiring admission to a divisional hospital PICU in May 2016 compared to May 2013-2015 (Incidence rate ratio: 1.7 [95% CI: 1.1-2.6]). This increase was not observed in children aged 5-14 years. The case-fatality ratio was not significantly different in 2016 compared to previous years. CONCLUSION: The investigation enabled targeted public health response measures, including enhanced SARI surveillance at divisional hospitals and an emergency influenza vaccination campaign in the Northern Division.

Response to the Ebola crisis in Sierra Leone.

The Ebola outbreak in 2014 marked the first time that an epidemic of this viral haemorrhagic fever had occurred in West Africa. From its origin in Guinea, the outbreak rapidly increased to become a humanitarian crisis affecting all aspects of life in the three countries worst affected: Guinea, Sierra Leone and Liberia. Improving understanding of Ebola virus disease among the general population and instigating the behavioural changes required to help break the epidemic were central to the public health response. This article explores some of the misconceptions about Ebola as it spread into Sierra Leonean communities, and the social mobilisation response of the government of Sierra Leone. It is a reflective account of conversations with Sierra Leonean nationals during a military deployment at the International Security Advisory Team headquarters medical treatment facility in Freetown.

Correlation of clinical trachoma and infection in Aboriginal communities.

BACKGROUND: Trachoma is the leading infectious cause of blindness due to conjunctival infection with Chlamydia trachomatis. The presence of active trachoma and evidence of infection are poorly correlated and a strong immunologically-mediated inflammatory response means that clinical signs last much longer than infection. This population-based study in five Aboriginal communities endemic for trachoma in northern Australia compared a fine grading of clinical trachoma with diagnostic positivity and organism load. METHODS: A consensus fine grading of trachoma, based on clinical assessment and photograding, was compared to PCR, a lipopolysacharide (LPS)-based point-of-care (POC) and a 16S RNA-based nucleic acid amplification test (NAAT). Organism load was measured in PCR positive samples. RESULTS: A total of 1282 residents, or 85.2% of the study population, was examined. Taking the findings of both eyes, the prevalence of trachomatous inflammation-follicular (TF) in children aged 1-9 years was 25.1% (96/383) of whom 13 (13.7%) were PCR positive on the left eye. When clinical data were limited to the left eye as this was tested for PCR, the prevalence of TF decreased to 21.4% (82/383). The 301 TF negative children, 13 (4.3%) were PCR positive. The fine grading of active trachoma strongly correlated with organism load and disease severity (rs = 0.498, P = 0.0004). Overall, 53% of clinical activity (TF(1) or TF(2)) and 59% of PCR positivity was found in those with disease scores less than the WHO simplified grade of TF. CONCLUSION: Detailed studies of the pathogenesis, distribution and natural history of trachoma should use finer grading schemes for the more precise identification of clinical status. In low prevalence areas, the LPS-based POC test lacks the sensitivity to detect active ocular infection and nucleic acid amplification tests such as PCR or the 16S-RNA based NAAT performed better. Trachoma in the Aboriginal communities requires specific control measures.

Severe pertussis in infants: estimated impact of first vaccine dose at 6 versus 8 weeks in australia.

We estimated the potential benefits of advancing the first dose of pertussis vaccine for infants from 8 to 6 weeks of age, using Australian national disease databases. Infants had notification rates 3-fold greater than the general population and accounted for 52% of recorded hospitalizations. Infants 1 and 2 months of age had notification rates 3.5 times (95% CI: 2.7-4.5) higher than infants 3 to 11 months of age. Estimation of acceleration of the vaccine to 6 weeks of age reduced average notifications, hospitalizations, and hospital bed-days by 8%, 9%, and 12%, respectively, with larger reductions in an epidemic year.

Attempts at producing a hybridised Penaeus mondon cell line by cellular fusion.

The lack of a standardised system for the isolation, identification and purification of prawn viruses, is a major obstacle to the control of viruses in penaeid aquaculture. To date, spontaneous and induced transformation of somatic penaeid cells has failed. Hybrid cells with the aim of supporting the growth of penaeid viruses were created using polyethylene glycol (PEG)-mediated fusion with two immortal cell lines, Epithelioma papulosum cyprinid (EPC) and Spodoptera frugiperda pupal ovarian cells (Sf9), fused with Penaeus monodon haemocytes. The immortal cell lines were biochemically blocked with actinomycin D and puromycin before fusion occurred. A total of 78 hybrid clones were created. The methods used to confirm the presence of P. monodon genes and proteins in the hybrid cells did not detect crustacean components, nor was any viral amplification detected by real-time PCR after hybrid cells were inoculated with two P. monodon parvoviruses, Penaeus merguiensis densovirus and infectious hypodermal and haematopoietic necrosis virus. These results suggest although the creation of the hybrid cells appeared successful, the cell lines lacked key crustacean cell components required for their use as an in vitro system for virus replication.

Australia's notifiable disease status, 2007: annual report of the National Notifiable Diseases Surveillance System.

In 2007, 69 diseases and conditions were nationally notifiable in Australia. States and territories reported a total of 146,991 notifications of communicable diseases to the National Notifiable Diseases Surveillance System, an increase of 5% on the number of notifications in 2006. In 2007, the most frequently notified diseases were sexually transmissible infections (62,474 notifications, 43% of total notifications), gastrointestinal diseases (30,325 notifications, 21% of total notifications) and vaccine preventable diseases (25,347 notifications, 17% of total notifications). There were 19,570 notifications of bloodborne diseases; 6,823 notifications of vectorborne diseases; 1,762 notifications of other bacterial infections; 687 notifications of zoonoses and 3 notifications of quarantinable diseases.