Browsing by Author "Ahmed S"

Now showing 1 - 5 of 5

Early user perspectives on using computer-aided detection software for interpreting chest X-ray images to enhance access and quality of care for persons with tuberculosis.
(2023-Dec-21) Creswell J; Vo LNQ; Qin ZZ; Muyoyeta M; Tovar M; Wong EB; Ahmed S; Vijayan S; John S; Maniar R; Rahman T; MacPherson P; Banu S; Codlin AJ
Despite 30 years as a public health emergency, tuberculosis (TB) remains one of the world's deadliest diseases. Most deaths are among persons with TB who are not reached with diagnosis and treatment. Thus, timely screening and accurate detection of TB, particularly using sensitive tools such as chest radiography, is crucial for reducing the global burden of this disease. However, lack of qualified human resources represents a common limiting factor in many high TB-burden countries. Artificial intelligence (AI) has emerged as a powerful complement in many facets of life, including for the interpretation of chest X-ray images. However, while AI may serve as a viable alternative to human radiographers and radiologists, there is a high likelihood that those suffering from TB will not reap the benefits of this technological advance without appropriate, clinically effective use and cost-conscious deployment. The World Health Organization recommended the use of AI for TB screening in 2021, and early adopters of the technology have been using the technology in many ways. In this manuscript, we present a compilation of early user experiences from nine high TB-burden countries focused on practical considerations and best practices related to deployment, threshold and use case selection, and scale-up. While we offer technical and operational guidance on the use of AI for interpreting chest X-ray images for TB detection, our aim remains to maximize the benefit that programs, implementers, and ultimately TB-affected individuals can derive from this innovative technology.
Expanding molecular diagnostic coverage for tuberculosis by combining computer-aided chest radiography and sputum specimen pooling: a modeling study from four high-burden countries.
(2024) Codlin AJ; Vo LNQ; Garg T; Banu S; Ahmed S; John S; Abdulkarim S; Muyoyeta M; Sanjase N; Wingfield T; Iem V; Squire B; Creswell J
BACKGROUND: In 2022, fewer than half of persons with tuberculosis (TB) had access to molecular diagnostic tests for TB due to their high costs. Studies have found that the use of artificial intelligence (AI) software for chest X-ray (CXR) interpretation and sputum specimen pooling can each reduce the cost of testing. We modeled the combination of both strategies to estimate potential savings in consumables that could be used to expand access to molecular diagnostics. METHODS: We obtained Xpert testing and positivity data segmented into deciles by AI probability scores for TB from the community- and healthcare facility-based active case finding conducted in Bangladesh, Nigeria, Viet Nam, and Zambia. AI scores in the model were based on CAD4TB version 7 (Zambia) and qXR (all other countries). We modeled four ordinal screening and testing approaches involving AI-aided CXR interpretation to indicate individual and pooled testing. Setting a false negative rate of 5%, for each approach we calculated additional and cumulative savings over the baseline of universal Xpert testing, as well as the theoretical expansion in diagnostic coverage. RESULTS: In each country, the optimal screening and testing approach was to use AI to rule out testing in deciles with low AI scores and to guide pooled vs individual testing in persons with moderate and high AI scores, respectively. This approach yielded cumulative savings in Xpert tests over baseline ranging from 50.8% in Zambia to 57.5% in Nigeria and 61.5% in Bangladesh and Viet Nam. Using these savings, diagnostic coverage theoretically could be expanded by 34% to 160% across the different approaches and countries. CONCLUSIONS: Using AI software data generated during CXR interpretation to inform a differentiated pooled testing strategy may optimize TB diagnostic test use, and could extend molecular tests to more people who need them. The optimal AI thresholds and pooled testing strategy varied across countries, which suggests that bespoke screening and testing approaches may be needed for differing populations and settings. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s44263-024-00081-2.
Neonatal mortality risk of vulnerable newborns by fine stratum of gestational age and birthweight for 230 679 live births in nine low- and middle-income countries, 2000-2017.
(2024-Jan-16) Hazel EA; Erchick DJ; Katz J; Lee ACC; Diaz M; Wu LSF; West KP; Shamim AA; Christian P; Ali H; Baqui AH; Saha SK; Ahmed S; Roy AD; Silveira MF; Buffarini R; Shapiro R; Zash R; Kolsteren P; Lachat C; Huybregts L; Roberfroid D; Zhu Z; Zeng L; Gebreyesus SH; Tesfamariam K; Adu-Afarwuah S; Dewey KG; Gyaase S; Poku-Asante K; Boamah Kaali E; Jack D; Ravilla T; Tielsch J; Taneja S; Chowdhury R; Ashorn P; Maleta K; Ashorn U; Mangani C; Mullany LC; Khatry SK; Ramokolo V; Zembe-Mkabile W; Fawzi WW; Wang D; Schmiegelow C; Minja D; Msemo OA; Lusingu JPA; Smith ER; Masanja H; Mongkolchati A; Keentupthai P; Kakuru A; Kajubi R; Semrau K; Hamer DH; Manasyan A; Pry JM; Chasekwa B; Humphrey J; Black RE
OBJECTIVE: To describe the mortality risks by fine strata of gestational age and birthweight among 230 679 live births in nine low- and middle-income countries (LMICs) from 2000 to 2017. DESIGN: Descriptive multi-country secondary data analysis. SETTING: Nine LMICs in sub-Saharan Africa, Southern and Eastern Asia, and Latin America. POPULATION: Liveborn infants from 15 population-based cohorts. METHODS: Subnational, population-based studies with high-quality birth outcome data were invited to join the Vulnerable Newborn Measurement Collaboration. All studies included birthweight, gestational age measured by ultrasound or last menstrual period, infant sex and neonatal survival. We defined adequate birthweight as 2500-3999 g (reference category), macrosomia as ≥4000 g, moderate low as 1500-2499 g and very low birthweight as <1500 g. We analysed fine strata classifications of preterm, term and post-term: ≥42 MAIN OUTCOME MEASURES: Median and interquartile ranges by study for neonatal mortality rates (NMR) and relative risks (RR). We also performed meta-analysis for the relative mortality risks with 95% confidence intervals (CIs) by the fine categories, stratified by regional study setting (sub-Saharan Africa and Southern Asia) and study-level NMR (≤25 versus >25 neonatal deaths per 1000 live births). RESULTS: We found a dose-response relationship between lower gestational ages and birthweights with increasing neonatal mortality risks. The highest NMR and RR were among preterm babies born at <28 weeks (median NMR 359.2 per 1000 live births; RR 18.0, 95% CI 8.6-37.6) and very low birthweight (462.8 per 1000 live births; RR 43.4, 95% CI 29.5-63.9). We found no statistically significant neonatal mortality risk for macrosomia (RR 1.1, 95% CI 0.6-3.0) but a statistically significant risk for all preterm babies, post-term babies (RR 1.3, 95% CI 1.1-1.5) and babies born at 37 CONCLUSIONS: In addition to tracking vulnerable newborn types, monitoring finer categories of birthweight and gestational age will allow for better understanding of the predictors, interventions and health outcomes for vulnerable newborns. It is imperative that all newborns from live births and stillbirths have an accurate recorded weight and gestational age to track maternal and neonatal health and optimise prevention and care of vulnerable newborns.
Neonatal mortality risk of vulnerable newborns: A descriptive analysis of subnational, population-based birth cohorts for 238 203 live births in low- and middle-income settings from 2000 to 2017.
(2023-May-08) Hazel EA; Erchick DJ; Katz J; Lee ACC; Diaz M; Wu LSF; West KP; Shamim AA; Christian P; Ali H; Baqui AH; Saha SK; Ahmed S; Roy AD; Silveira MF; Buffarini R; Shapiro R; Zash R; Kolsteren P; Lachat C; Huybregts L; Roberfroid D; Zhu Z; Zeng L; Gebreyesus SH; Tesfamariam K; Adu-Afarwuah S; Dewey KG; Gyaase S; Poku-Asante K; Boamah Kaali E; Jack D; Ravilla T; Tielsch J; Taneja S; Chowdhury R; Ashorn P; Maleta K; Ashorn U; Mangani C; Mullany LC; Khatry SK; Ramokolo V; Zembe-Mkabile W; Fawzi WW; Wang D; Schmiegelow C; Minja D; Msemo OA; Lusingu JPA; Smith ER; Masanja H; Mongkolchati A; Keentupthai P; Kakuru A; Kajubi R; Semrau K; Hamer DH; Manasyan A; Pry JM; Chasekwa B; Humphrey J; Black RE
OBJECTIVE: We aimed to understand the mortality risks of vulnerable newborns (defined as preterm and/or born weighing smaller or larger compared to a standard population), in low- and middle-income countries (LMICs). DESIGN: Descriptive multi-country, secondary analysis of individual-level study data of babies born since 2000. SETTING: Sixteen subnational, population-based studies from nine LMICs in sub-Saharan Africa, Southern and Eastern Asia, and Latin America. POPULATION: Live birth neonates. METHODS: We categorically defined five vulnerable newborn types based on size (large- or appropriate- or small-for-gestational age [LGA, AGA, SGA]), and term (T) and preterm (PT): T + LGA, T + SGA, PT + LGA, PT + AGA, and PT + SGA, with T + AGA (reference). A 10-type definition included low birthweight (LBW) and non-LBW, and a four-type definition collapsed AGA/LGA into one category. We performed imputation for missing birthweights in 13 of the studies. MAIN OUTCOME MEASURES: Median and interquartile ranges by study for the prevalence, mortality rates and relative mortality risks for the four, six and ten type classification. RESULTS: There were 238 203 live births with known neonatal status. Four of the six types had higher mortality risk: T + SGA (median relative risk [RR] 2.6, interquartile range [IQR] 2.0-2.9), PT + LGA (median RR 7.3, IQR 2.3-10.4), PT + AGA (median RR 6.0, IQR 4.4-13.2) and PT + SGA (median RR 10.4, IQR 8.6-13.9). T + SGA, PT + LGA and PT + AGA babies who were LBW, had higher risk compared with non-LBW babies. CONCLUSIONS: Small and/or preterm babies in LIMCs have a considerably increased mortality risk compared with babies born at term and larger. This classification system may advance the understanding of the social determinants and biomedical risk factors along with improved treatment that is critical for newborn health.
Vulnerable newborn types: analysis of subnational, population-based birth cohorts for 541 285 live births in 23 countries, 2000-2021.
(2023-May-08) Erchick DJ; Hazel EA; Katz J; Lee ACC; Diaz M; Wu LSF; Yoshida S; Bahl R; Grandi C; Labrique AB; Rashid M; Ahmed S; Roy AD; Haque R; Shaikh S; Baqui AH; Saha SK; Khanam R; Rahman S; Shapiro R; Zash R; Silveira MF; Buffarini R; Kolsteren P; Lachat C; Huybregts L; Roberfroid D; Zeng L; Zhu Z; He J; Qiu X; Gebreyesus SH; Tesfamariam K; Bekele D; Chan G; Baye E; Workneh F; Asante KP; Kaali EB; Adu-Afarwuah S; Dewey KG; Gyaase S; Wylie BJ; Kirkwood BR; Manu A; Thulasiraj RD; Tielsch J; Chowdhury R; Taneja S; Babu GR; Shriyan P; Ashorn P; Maleta K; Ashorn U; Mangani C; Acevedo-Gallegos S; Rodriguez-Sibaja MJ; Khatry SK; LeClerq SC; Mullany LC; Jehan F; Ilyas M; Rogerson SJ; Unger HW; Ghosh R; Musange S; Ramokolo V; Zembe-Mkabile W; Lazzerini M; Rishard M; Wang D; Fawzi WW; Minja DTR; Schmiegelow C; Masanja H; Smith E; Lusingu JPA; Msemo OA; Kabole FM; Slim SN; Keentupthai P; Mongkolchati A; Kajubi R; Kakuru A; Waiswa P; Walker D; Hamer DH; Semrau KEA; Chaponda EB; Chico RM; Banda B; Musokotwane K; Manasyan A; Pry JM; Chasekwa B; Humphrey J; Black RE
OBJECTIVE: To examine prevalence of novel newborn types among 541 285 live births in 23 countries from 2000 to 2021. DESIGN: Descriptive multi-country secondary data analysis. SETTING: Subnational, population-based birth cohort studies (n = 45) in 23 low- and middle-income countries (LMICs) spanning 2000-2021. POPULATION: Liveborn infants. METHODS: Subnational, population-based studies with high-quality birth outcome data from LMICs were invited to join the Vulnerable Newborn Measurement Collaboration. We defined distinct newborn types using gestational age (preterm [PT], term [T]), birthweight for gestational age using INTERGROWTH-21st standards (small for gestational age [SGA], appropriate for gestational age [AGA] or large for gestational age [LGA]), and birthweight (low birthweight, LBW [<2500 g], nonLBW) as ten types (using all three outcomes), six types (by excluding the birthweight categorisation), and four types (by collapsing the AGA and LGA categories). We defined small types as those with at least one classification of LBW, PT or SGA. We presented study characteristics, participant characteristics, data missingness, and prevalence of newborn types by region and study. RESULTS: Among 541 285 live births, 476 939 (88.1%) had non-missing and plausible values for gestational age, birthweight and sex required to construct the newborn types. The median prevalences of ten types across studies were T+AGA+nonLBW (58.0%), T+LGA+nonLBW (3.3%), T+AGA+LBW (0.5%), T+SGA+nonLBW (14.2%), T+SGA+LBW (7.1%), PT+LGA+nonLBW (1.6%), PT+LGA+LBW (0.2%), PT+AGA+nonLBW (3.7%), PT+AGA+LBW (3.6%) and PT+SGA+LBW (1.0%). The median prevalence of small types (six types, 37.6%) varied across studies and within regions and was higher in Southern Asia (52.4%) than in Sub-Saharan Africa (34.9%). CONCLUSIONS: Further investigation is needed to describe the mortality risks associated with newborn types and understand the implications of this framework for local targeting of interventions to prevent adverse pregnancy outcomes in LMICs.