Feasibility and early clinical yield of handheld AI-assisted obstetric point-of-care ultrasound in routine antenatal care: a pilot study in Addis Ababa, Ethiopia

Feasibility and early clinical yield of handheld AI-assisted obstetric point-of-care ultrasound in routine antenatal care: a pilot study in Addis Ababa, Ethiopia | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Feasibility and early clinical yield of handheld AI-assisted obstetric point-of-care ultrasound in routine antenatal care: a pilot study in Addis Ababa, Ethiopia Crystal Richardson, Messay Gebrekidan, Melkamu Hunegnaw Asmare, and 6 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9101106/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background: The World Health Organization recommends at least one ultrasound examination before 24 weeks of gestation to improve gestational-age assessment and detection of multiple pregnancy and fetal anomalies. However, access to obstetric ultrasound at the primary-care level remains limited in many low-resource settings, including Ethiopia. This study evaluated the feasibility and early clinical yield of integrating AI-enabled handheld obstetric point-of-care ultrasound (POCUS) into routine antenatal care in four government health centers in Addis Ababa. Methods: We conducted a 4-week prospective pilot across four urban health centers (Abuare, Signal-Woreda 7, Kebena, and Janmeda). Pregnant women attending routine antenatal care (ANC) underwent focused obstetric POCUS using Butterfly handheld devices. Trained frontline providers, mainly junior GPs and midwives, performed scans using a standardized acquisition form, while clinical data were entered into REDCap and linked to archived image records on the Butterfly cloud platform. Primary outcomes were the feasibility of workflow integration, completeness of focused scan variables, the spectrum of focused obstetric findings, and referrals generated. Results: A total of 101 women were scanned. Mean maternal age was 28.3 years (SD 4.4); median gravidity was 2 (IQR 1–3), and median parity was 1 (IQR 0–2). Ultrasound-estimated gestational age was recorded in 100/101 women (median 30 weeks; IQR 16.8–33.0). Fetal cardiac activity was present in 92/99 (92.9%), absent in 4/99 (4.0%), and indeterminate in 3/99 (3.0%). Fetal presentation was cephalic in 65/101 (64.4%), breech in 5/101 (5.0%), transverse in 17/101 (16.8%), and indeterminate in 14/101 (13.9%). Placental location was low-lying in 1/100 (1.0%), and qualitative amniotic fluid was low in 1/100 (1.0%). Seven of 99 women (7.1%) were referred to a higher-level facility. A binary normal-scan indicator classified 93/101 scans (92.1%) as normal. Conclusions: Handheld focused obstetric POCUS was operationally feasible within routine urban ANC and identified a clinically relevant minority of women requiring referral or further review. Larger studies with standardized expert over-read, referral follow-up, and explicit image-quality metrics are needed before stronger claims about referral optimization or broader clinical impact can be made. Maternal & Fetal Medicine Biomedical Engineering Artificial Intelligence and Machine Learning Obstetric POCUS Point-of-care ultrasound Antenatal care Ethiopia Butterfly ultrasound Primary health care Implementation Feasibility Figures Figure 1 Figure 2 Figure 3 Background Ultrasound is now a core component of modern antenatal care. The World Health Organization recommends one ultrasound examination before 24 weeks’ gestation to improve gestational-age estimation, detect multiple pregnancy and fetal anomalies earlier, reduce post-term induction, and improve the pregnancy experience [ 1 ]. At the same time, WHO guidance does not support routine fetal Doppler as a standard population-level ANC intervention for all pregnancies [ 1 ]. Despite that recommendation, access to timely obstetric ultrasound remains uneven across sub-Saharan Africa, where imaging is still often concentrated in hospitals and specialist settings. Recent African implementation studies show that task-shared obstetric POCUS can be delivered in low-resource settings when training, mentorship, supervision, and quality assurance are built into the model. In Kenya, large-scale training and later RE-AIM evaluation showed that scale-up is possible but depends on ongoing mentorship, maintenance, and implementation support [ 2 , 3 ]. In Malawi, midwife-delivered routine ANC POCUS was highly acceptable and achieved acceptable image quality in most reviewed scans [ 4 ]. Ethiopian evidence already supports the value of decentralized portable obstetric ultrasound. A retrospective study from semi-urban health centers reported clinically important abnormal findings in 12.7% of scanned pregnancies and referral of 98.4% of those abnormal cases [ 5 ], while a later pre-post study found that institutionalizing limited obstetric ultrasound was associated with improved ANC, skilled delivery, and postnatal service utilization [ 6 ]. However, published Ethiopian work has provided less detail on digitally linked image archiving, specialist over-read, and structured quality assurance within routine urban government primary-care workflows. Against this background, our pilot evaluated the structured integration of handheld obstetric POCUS into routine ANC across four urban government health centers in Addis Ababa. The purpose was not to replace formal radiology, but to assess whether a focused handheld ultrasound package could be embedded into primary-care ANC workflows, generate usable focused findings, support referral decisions, and operate within a digitally linked quality-assurance pathway. We aimed to evaluate the feasibility of workflow integration, the focused obstetric findings patterns, and the referral implications of handheld obstetric POCUS embedded in routine ANC at four urban primary health centers in Addis Ababa. Methods Study design and setting This study was designed as a prospective, multi-center feasibility and implementation pilot to evaluate the integration of handheld AI-enabled obstetric point-of-care ultrasound (POCUS) into routine antenatal care (ANC) services at the primary health care level in Addis Ababa, Ethiopia. Butterfly Network is approved in Ethiopia but used in a few facilities, with limited exposure outside major teaching hospitals. The pilot was conducted over a fixed 4-week implementation period in four urban government health centers: Abuare, Signal-Woreda 7, Kebena, and Janmeda, following submission and institutional review of an integrated obstetric POCUS proposal and in coordination with local clinical radiology leadership and relevant health authorities. Rather than establishing a stand-alone research clinic, the intervention was embedded within existing ANC workflows to assess real-world operational feasibility, uptake, and clinical usefulness at the primary-care level. All the Butterfly OB POCUS was reported on the mother’s record indicating effective and comparative outcome. The pilot was conceived as an implementation-oriented service integration study rather than an efficacy trial. The primary purpose was to determine whether a structured package of focused obstetric ultrasound, supported by standardized documentation, digital image storage, and specialist oversight, could be operationalized within routine ANC consultations in urban health centers. The study also sought to characterize the pattern of focused obstetric findings generated in this setting and to document whether scan findings led to continued management at the health center or referral to a higher level of care. Participants and eligibility All consecutive pregnant women presenting for routine ANC at participating health centers during the study period were considered for inclusion. Eligibility was intentionally broad because the pilot aimed to evaluate service integration under routine practice conditions rather than under highly restricted trial criteria. Women attending ANC who agreed to undergo focused obstetric POCUS as part of the visit were enrolled, allowing the study to capture the case mix typically encountered in primary-care ANC, including both early pregnancy viability/dating encounters and later gestation follow-up visits. Participation was voluntary, and women who declined an ultrasound continued to receive standard ANC without affecting their routine care. At enrollment, standardized baseline information was collected using structured study tools. Variables recorded at the point of care included maternal age, gravidity, parity, gestational age by last menstrual period when available, prior obstetric history, ANC follow-up status, maternal education level, residence type, and predefined high-risk clinical indicators such as vaginal bleeding, abdominal pain, reduced fetal movement, or hypertensive symptoms. Immediate clinical disposition after scanning, including continued management at the primary-care facility or referral to a higher-level facility, was also documented. The analytic dataset ultimately included 101 enrolled women, each linked to a unique study identifier. Intervention and training The intervention consisted of a structured introduction of handheld obstetric POCUS into routine ANC using Butterfly ultrasound devices connected to secure study mobile phones, standardized focused scan forms, defined data-entry procedures, and consultant Radiologist specialist-supported oversight. Service delivery was based on a task-shared model involving five junior general practitioners together with site-based midwives, with two midwives participating at each health center. This model was selected to reflect the existing primary maternal health workforce and to test whether focused obstetric ultrasound could be incorporated into routine care without creating a parallel imaging service. Before implementation, participating providers underwent institutional orientation and focused practical training in core obstetric POCUS applications relevant to primary care. Training covered four linked domains: image acquisition, focused interpretation within scope, standardized documentation on the obstetric POCUS form, and escalation of uncertain or abnormal findings through the specialist review pathway. Three Butterfly handheld ultrasound devices were deployed across the four participating sites and connected to secure mobile phones with administrator-controlled access. Focused scan package The ultrasound protocol used in this pilot was a focused obstetric POCUS package rather than a comprehensive formal obstetric sonography examination. The structured acquisition form captured the following core domains at the time of scanning: fetal cardiac activity, gestational age by ultrasound, fetal presentation, placental location, qualitative assessment of amniotic fluid, gross fetal abnormality screening, and Doppler assessment. In clinically relevant early pregnancy cases, the focused examination also supported confirmation of intrauterine pregnancy and assessment of viability concerns. The focused package was designed to support clinically actionable bedside decision-making in ANC rather than to replace formal radiology or full fetal anomaly scanning. Accordingly, fetal presentation, placental location, and amniotic fluid were recorded as focused screening variables, while “gross abnormality suspected” functioned as a trigger for further review rather than as a definitive diagnostic classification. Data capture and quality assurance Data capture used a linked clinical-imaging workflow designed to preserve standardization while allowing case-level review. At the point of care, maternal characteristics and focused ultrasound findings were recorded using a standardized obstetric POCUS data collection tool. Ultrasound images were then stored in the secure Butterfly cloud-based application using administrator-controlled accounts, while corresponding clinical and demographic variables were entered into REDCap using unique study identifiers. This architecture enabled de-identified linkage between scan findings and stored images without exporting direct patient identifiers into the analytic dataset. The REDCap dataset used study IDs ranging from 1 to 101. The Butterfly iQ3 ultrasound system includes automated iQ Fan and iQ Slice tools, multiple imaging modes, and over 20 anatomical presets, enabling comprehensive obstetric scanning, looking at details, effective umbilical doppler assessment, high image resolution, and facilitating easy training of research assistants during the pilot. To improve data integrity, REDCap fields were structured in predefined columns and incorporated validation rules and range checks. Access to the Butterfly platform and REDCap database was restricted to authorized study personnel, HIPAA-regulated, and only de-identified data were exported for analysis. The pilot also incorporated a quality-assurance pathway through specialist-supported oversight, with scanning performed under radiologist supervision, creating opportunities for teleconsultation and expert confirmation of findings. Outcomes The study outcomes were specified to capture both the implementation feasibility of integrating handheld obstetric POCUS into routine antenatal care and the early clinical yield of the focused scan package in a primary health care setting. Primary outcomes included: (i) feasibility of routine ANC integration, defined as successful incorporation of focused obstetric POCUS into the existing ANC workflow at each of the four participating health centers during the pilot period, without the need to establish a parallel imaging service; (ii) completeness of focused scan fields, defined by the proportion of scans for which the core structured variables were recorded in REDCap; (iii) proportion of scans classified as interpretable versus indeterminate, using the structured recording options for variables marked as “unable to determine,” “uncertain,” or “not diagnostic”; (iv) focused finding spectrum, defined as the distribution of principal ultrasound findings identified during the pilot; and (v) referrals generated, defined as the number and proportion of women whose scan findings resulted in referral from the health center to a higher-level facility. Secondary outcomes included the proportion of women receiving their first-ever ultrasound in the current pregnancy, the overall abnormality detection rate, the radiologist over-read concordance rate, final expert-reviewed abnormality categories, and site-level variation in scan volume, completeness, finding patterns, indeterminate scan fields, and referral frequency across the participating health centers. Secondary outcomes also included accuracy and completeness of AI-assisted fetal biometric measurements used for gestational age estimation and variation in clinical detection patterns across trimesters. Statistical analysis Because this was a feasibility pilot, analyses were primarily descriptive. Continuous variables were summarized using median (IQR) and mean (SD), and categorical variables using counts and percentages with non-missing denominators. Exact 95% confidence intervals were calculated for key proportions. No formal power calculation was performed, and any site-level comparisons were considered exploratory. Butterfly IQ 3 ultrasound was AI-powered, and most biometric calculations were automatic. Ethics Ethical approval was obtained from the relevant institutional body. Written informed consent was obtained from all participants before enrollment. Participation was voluntary, and women retained the right to decline an ultrasound without any effect on their routine ANC care. Ultrasound images were stored on the secure Butterfly cloud platform, and clinical data were entered into password-protected REDCap databases using de-identified study IDs. Access was restricted to authorized study personnel, and no directly identifiable patient information was included in analytic exports. Results Cohort and site distribution A total of 101 pregnant women underwent focused obstetric POCUS during the 4-week pilot across four government health centers in Addis Ababa. Site distribution was Signal-Woreda 7, 37/101 (36.6%); Janmeda, 33/101 (32.7%); Abuare, 22/101 (21.8%); and Kebena, 9/101 (8.9%). Mean maternal age was 28.3 years (SD 4.4; range 18–39). Median gravidity was 2 (IQR 1–3), and median parity was 1 (IQR 0–2). Among participants with residence data, 98/101 (99.0%) were urban residents. The highest education level was primary in 38/101 (38.0%), secondary in 30/100 (30.0%), and college in 31/100 (31.0%), with 1/100 (1.0%) reporting no formal education. Full participant characteristics are presented in Table 1. Gestational-age and viability profile Ultrasound-estimated gestational age was recorded for 100/101 women (1 missing). The median gestational age was 30 weeks (IQR 16.8–33.0), with a mean of 25.7 weeks (SD 10.3; range 0–39). By trimester distribution, the first trimester accounted for approximately 19/101, the second trimester for approximately 24/101, and the third trimester for approximately 58/101 scans. Fetal cardiac activity was documented as present in 92/99 scans (92.9%), absent in 4/99 (4.0%) fetal demises, and indeterminate (unable to determine) in 3/99 (3.0%), mostly in early 1st trimester pregnancies, with 2 values missing. Focused obstetric findings Across all scans, fetal presentation was recorded as cephalic in 65/101 (64.4%), breech in 5/101 (5.0%), transverse in 17/101 (16.8%), and unable to determine in 14/101 (13.9%). Placental location was anterior in 54/100 (54.0%), posterior in 22/100 (22.0%), fundal in 16/100 (16.0%), low-lying in 1/100 (1.0%), and uncertain in 7/100 (7.0%), with 1 value missing. Uncertain placental location was predominantly reported in early first-trimester pregnancies. Qualitative amniotic fluid was normal in 95/100 (95.0%), low in 1/100 (1.0%), and uncertain in 4/100 (4.0%), with 1 value missing and no cases of high fluid. One congenital uterine abnormality subseptate uterus, was detected (Fig 3, Panel 3 ). Complete focused obstetric POCUS findings are detailed in Table 2. In one case, a mother at 37 weeks + 3 days gestation was initially diagnosed with anhydramnios and recommended for emergency cesarean section by MFM obstetrician scan. Re-evaluation using Butterfly POCUS, in consultation with a radiologist, identified a posterior amniotic fluid pocket measuring 11 cm with a reassuring biophysical profile (BPP). The mother was subsequently followed and later had a safe vaginal delivery. Referral and non-normal findings A binary normal-scan indicator classified 93/101 scans (92.1%) as normal and 8/101 (7.9%) as non-normal. Seven of 99 women (7.1%) were referred following the scan encounter, with 2 values missing for referral status. Site-level implementation outcomes are presented in Table 3. Table 1. Participant characteristics (N = 101) Characteristic Value N (non-missing) Age (years), mean ± SD 28.3 ± 4.4 101 Range 18–39 Gravidity, median (IQR) 2 (1–3) 101 Range 1–6 Parity, median (IQR) 1 (0–2) 101 Range 0–4 Highest education, n/N (%) 100 None 1/100 (1.0%) Primary 38/100 (38.0%) Secondary 30/100 (30.0%) College 31/100 (31.0%) Residence, n/N (%) 99 Urban 98/99 (99.0%) Peri-urban 1/99 (1.0%) Health center site, n/N (%) 101 Woreda 7 37/101 (36.6%) Janmeda 33/101 (32.7%) Abuare 22/101 (21.8%) Kebena 9/101 (8.9%) SD, standard deviation; IQR, interquartile range. Table 2. Focused obstetric POCUS findings (N = 101) Finding n/N (%) Missing Fetal cardiac activity 2 (2.0%) Present 92/99 (92.9%) Absent 4/99 (4.0%) Unable to determine 3/99 (3.0%) Gestational age (weeks) 1 (1.0%) Mean ± SD 25.7 ± 10.3 Median (IQR) 30 (16.8–33.0) Calculated trimester 0 (0.0%) First (≤13 weeks) 19/101 (18.8%)† Second (14–27 weeks) 24/101 (23.8%)† Third (≥28 weeks) 58/101 (57.4%)† Fetal presentation 0 (0.0%) Cephalic 69/101 (68.3%) Breech 5/101 (5.0%) Transverse 17/101 (16.8%) Unable to determine 10/101 (9.9%) Placental location 1 (1.0%) Anterior 54/100 (54.0%) Posterior 22/100 (22.0%) Fundal 16/100 (16.0%) Low-lying 1/100 (1.0%) Uncertain 7/100 (7.0%) Amniotic fluid (qualitative) 1 (1.0%) Normal 95/100 (95.0%) Low 1/100 (1.0%) High 0/100 (0.0%) Uncertain 4/100 (4.0%) Normal scan indicator 0 (0.0%) Normal (1) 93/101 (92.1%) Non-normal (0) 8/101 (7.9%) Referral made 2 (2.0%) Yes 7/99 (7.1%) No 92/99 (92.9%) † Trimester proportions derived from calculated trimester field (mean 2.44, SD 0.74). Percentages are approximate and may vary slightly based on gestational age coding. Table 3. Site-level implementation outcomes Indicator Signal-Woreda 7 Janmeda Abuare Kebena Total scans, n (%) 37 (36.6) 33 (32.7) 22 (21.8) 9 (8.9) Complete forms, n (%) 37 (100) 33 (100) 22 (100) 9 (100) Indeterminate presentation, n 7 1 1 1 Uncertain placental location, n 2 1 1 1 Uncertain amniotic fluid, n 2 1 1 0 Abnormal scans, n 3 3 2 0 Referrals, n 2 3 2 0 * NB-Most undetermined fetal presentations were recorded by GPs in the second trimester, when the fetus is mobile and placental location may be less clearly defined. Two cases of oligohydramnios and two fetal demises were categorized as undetermined AFI. Nearly all scans reviewed by the supervising radiologist demonstrated clear biometrics and definitive assessment of placental location, AFI, and fetal presentation. Discussion In this 4-week pilot, handheld focused obstetric POCUS was integrated into routine ANC across four urban government health centers in Addis Ababa, with 101 women scanned and low missingness across most core variables. Most scans were classified as normal, but a clinically meaningful minority showed absent or indeterminate fetal cardiac activity, non-cephalic presentation, abnormal or uncertain placental or fluid findings, or required referral. These findings support the operational feasibility of embedding a focused handheld ultrasound package in primary-care ANC, while also showing the importance of capturing indeterminate and uncertain results rather than masking them as missing data. The findings align with previous Ethiopian studies showing that portable obstetric ultrasound at the health-center level can detect important abnormalities and trigger referrals [ 5 ]. The novelty of this study lies in integrating handheld POCUS with REDCap data capture, cloud image archiving, and specialist-supported review, addressing prior quality-assurance gaps [ 5 ]. The broader African implementation literature strengthens this interpretation. In Kenya, large-scale programmatic rollout showed that non-radiologist obstetric POCUS can be scaled when training, supervision, and health-system support are sustained [ 2 , 3 ]. In Malawi, midwife-performed ANC POCUS was highly acceptable and achieved acceptable scan quality in most reviewed scans [ 4 ]. Together, these studies suggest that feasibility in low-resource settings depends less on the handheld device alone than on the surrounding implementation package: training, mentorship, image review, maintenance, consumables, and referral feedback. Access to imaging services plays a vital role in the diagnosis and treatment of a range of clinical conditions. However, low- and middle-income countries experience significant shortages in both imaging equipment and trained personnel [ 7 ]. POCUS has emerged as an important tool for improving access to diagnostic imaging in resource-limited primary care settings [ 8 ], with AI-enabled handheld devices such as the Butterfly iQ offering additional potential for standardized gestational-age estimation, anatomical scans, and decision support at the point of care [ 9 ]. The 2016 WHO antenatal care guidelines recommend at least one ultrasound before 24 weeks to estimate gestational age, detect fetal anomalies and multiple pregnancies, reduce post-term labor induction, and enhance the pregnancy experience [ 1 ]. AI-assisted functionality was integrated into the handheld POCUS platform to support fast image acquisition, automated fetal biometric measurements, and streamlined obstetric workflow even for beginners. AI protocols enabled simplified gestational age estimation through blind-sweep scanning and multi-slice imaging, while decision-support tools such as Fast Maternal and Fetal Ultrasound (FAMLI) guided evaluation of key obstetric conditions, including malpresentation, abnormal amniotic fluid, multiple gestation, and fetal viability [ 1 – 2 ]. Semi-automated documentation further improved reporting efficiency, supporting diagnostic consistency and safe task-shifting in primary care settings [ 2 – 3 ]. The implementation barriers reported elsewhere in Ethiopia and across LMIC settings are also relevant to this pilot. These include staff workload, shortages of trained personnel, service interruptions, consumable supply, maintenance, electricity, connectivity, and access to second-line review [ 10 ]. Recent Ethiopian tele-ultrasound experience suggests that remote specialist support can reduce travel burden and improve case escalation, making the Butterfly-plus-REDCap workflow in this study especially relevant to future scale-up models. Limitations This pilot study has limitations. It was conducted over a short 4-week period in four urban primary care centers with a small sample size (n = 101), limiting statistical power and the ability to draw firm conclusions regarding diagnostic accuracy, clinical outcomes, and long-term impact. The absence of a control group or pre–post comparison further limits causal interpretation. The study was performed in urban facilities with relatively reliable electricity and connectivity, which may not reflect conditions in rural settings. Limited handheld ultrasound devices and compatible mobile phones also constrained scan throughput at some sites. Although scans were supervised by a consultant radiologist and met quality standards, formal inter-observer testing and comparison with comprehensive radiology ultrasound were not performed, limiting assessment of diagnostic accuracy. Operational metrics and patient follow-up were also limited, and effective POCUS implementation requires structured training and supervision for primary healthcare providers. Future studies should include larger and more diverse populations, longer follow-up, and evaluation of diagnostic accuracy, referral completion, and maternal–neonatal outcomes. While the pilot demonstrates feasibility and early clinical yield, it does not yet establish improvements in outcomes, mortality, or referral appropriateness. Conclusions Handheld obstetric POCUS was feasibly integrated into routine antenatal care across four urban government health centers in Addis Ababa and identified clinically actionable findings requiring referral in a subset of women. Health center staff, including midwives and frontline health workers, demonstrated strong engagement and expressed a highly positive attitude toward the service, noting that the scans were carefully performed and valuable for patient care, and they supported its incorporation into routine practice. Implemented within Ministry-aligned governance, competency-based training, and consultant radiologist supervision, decentralized ultrasound shows promise for strengthening early detection of high-risk pregnancies at the primary care level. Larger prospective studies with standardized adjudication of abnormality, image-quality metrics, and referral follow-up are needed to confirm diagnostic performance, clinical impact, and scalability. Declarations Funding No specific funding was received for this study. Conflicts of interest The authors declare no conflicts of interest. Data availability Data will be co-owned by the primary investigators and securely stored on the Butterfly Cloud platform and the University of Washington REDCap system. De-identified data will be available upon reasonable ethical request. Authors’ contributions CR and MG conceptualized the study. The field work was coordinated by MG. CR, MG, DC, SH, EJ, AS, BW, and AS contributed to data collection. Data analysis and interpretation were conducted by MH, MG, and CR. All authors have read and approved the final document. Acknowledgements The authors thank all study participants, the volunteering GPs, midwives, and staff of the participating health centers for their support and collaboration. The investigators also acknowledge Riverstone Health for supplying the Butterfly handheld ultrasound devices used in this study, and the University of Washington for providing REDCap access for data management and structuring via the WWAMI Program. References World Health Organization. WHO Recommendations on Antenatal Care for a Positive Pregnancy Experience. Geneva: World Health Organization; 2016. Available from: https://www.who.int/publications/i/item/9789241549912 Wachira J, Matheka DM, Masheti SA, Githemo GK, Shah S, Haldeman MS, Ramos M, Bergman K. A training program for obstetrics point-of-care ultrasound to 514 rural healthcare providers in Kenya. BMC Med Educ. 2023;23(1). doi:10.1186/s12909-023-04886-x Githemo G, Wanyoro A, Masika J, Onsongo L, Bett S, Githuku S, Gachuiri G, Walker D, Santos N, Ghosh R, Otieno G. Evaluation of large-scale implementation of obstetric point-of-care ultrasound in eight counties in Kenya using RE-AIM framework. BMC Health Serv Res. 2025;25(1):1016. doi:10.1186/s12913-025-13212-8 Payesa C, Seyama L, Chimwaza Y, Sindani F, Kanise Y, Bvutula E, Phiri M, Nyangulu P, Gadama L, Kachale F, Gadama G, Mwale M, Yenokyan G, Sripad P, Hyre A, Noguchi LM, Dadabhai S. Feasibility and acceptability of point-of-care ultrasound delivered by midwives during routine antenatal care in Malawi: a prospective implementation science study. BMJ Open. 2025;15(8). doi:10.1136/bmjopen-2025-100515 Abawollo HS, Tsegaye ZT, Desta BF, Beshir IA, Mengesha BT, Guteta AA, Heyi AF, Mamo TT, Gebremedhin ZK, Damte HD, Zelealem M, Argaw MD. Contribution of portable obstetric ultrasound service innovation in averting maternal and neonatal morbidities and mortalities at semi-urban health centers of Ethiopia: a retrospective facility-based study. BMC Pregnancy Childbirth. 2022;22(1):368. doi:10.1186/s12884-022-04703-1 Woldeyohannes D, Abera M, Gebremedhin S, Tesfaye G. Institutionalization of limited obstetric ultrasound leading to increased antenatal, skilled delivery, and postnatal service utilization in three regions of Ethiopia: a pre-post study. PLoS One. 2023. doi:10.1371/journal.pone.0281626 Frija G, Blažić I, Frush DP, Hierath M, Kawooya M, Donoso-Bach L, Brkljačić B. How to improve access to medical imaging in low- and middle-income countries? eClinicalMedicine. 2021;38:101034. doi:10.1016/j.eclinm.2021.101034 Fleming KA, Horton S, Wilson ML, Atun R, DeStigter K, Flanigan J, et al. The Lancet Commission on diagnostics: transforming access to diagnostics. Lancet. 2021;398(10315):1997–2050. doi:10.1016/S0140-6736(21)00673-5 Della Ripa S, Santos N, Walker D. AI-enabled obstetric point-of-care ultrasound as an emerging technology in low- and middle-income countries: provider and health system perspectives. BMC Pregnancy Childbirth. 2025;25:729. doi:10.1186/s12884-025-07796-6 Ranger BJ, Bradburn E, Chen Q, Kim M, Noble JA, Papageorghiou AT. Portable ultrasound devices for obstetric care in resource-constrained environments: mapping the landscape. Gates Open Res. 2024;7:133. doi:10.12688/gatesopenres.15088.2 Additional Declarations The authors declare no competing interests. Supplementary Files Supplentarymaterials.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9101106","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":605074476,"identity":"c700b380-7537-4312-9b4d-345b739e4288","order_by":0,"name":"Crystal Richardson","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABEklEQVRIiWNgGAWjYDACCSB+AMRsDDxgkoGBvQFIGFjg15KAooXnAEiLBGEtDHAtEgkwceyAf3bzsw8JFYfz+djPHvxcUVYnZy75/OqGHwUSDPzt3QlYLblzzHhGwpnDlm08ecmSZ84dNracnVN2swfoMIkzZzdg02IgkWDMkNh224BNgsdAsrHtQOKG2zlpN3iAWgwkcnFoSf8M02L8s7GtLnHDzTNpN//g1ZIDt8UMaAtz4oYb7Mdu47NF4kZOMUPCmf8GbDw5ZpYNQL8YnMlhuy1jIMGDyy/8M9I3M3yoSDOQbz9jfLMBGGIGx48/u/nmj40cf3svVi3YAI8BmCRWOQiwPyBF9SgYBaNgFAx/AAAxWmAzsFTRpwAAAABJRU5ErkJggg==","orcid":"","institution":"RiverStone health","correspondingAuthor":true,"prefix":"","firstName":"Crystal","middleName":"","lastName":"Richardson","suffix":""},{"id":605074477,"identity":"56522bf9-d505-4de9-80c3-bc3a8d56ad70","order_by":1,"name":"Messay Gebrekidan","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA70lEQVRIiWNgGAWjYPACGyDmYTgA40oQoSUNpsWAaC2HwVoYiNKi297+8DPPr/OJ/f1nDx4u+PVHzpyB+eBtHjxazM6cMZbm7budOONGXsLhmX0GxpYNbMnWeLXcyGGQ5u25ndtwg8fgMG+PQeKGAzxm0ni13H/++Ddvz7nc+efPwLTwf8Ov5QYD0MwfB3I3HMgxOMzzA2wLG34tZ3LMLOc2JNdvvAHUwttgbGxwmM3Ycg4+LcePP77x5o+dsdz5M8afef7IyRkcb3544w0eLSDAxNsGZTGCGMwElIMV/vgDY/7Bp24UjIJRMApGKgAA+DtVo8g+pGYAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0001-6698-1321","institution":"TiruRad Diagnsotics and research center","correspondingAuthor":true,"prefix":"","firstName":"Messay","middleName":"","lastName":"Gebrekidan","suffix":""},{"id":605082669,"identity":"5926c73a-a0a3-4431-8401-8cc603dd4674","order_by":2,"name":"Melkamu Hunegnaw Asmare","email":"","orcid":"https://orcid.org/0000-0003-0990-3348","institution":"Leuven Center for Affordable Health Technology, KU Leuven, Andreas Vesaliusstraat 13, Leuven, 3000, Flanders, Belgium","correspondingAuthor":false,"prefix":"","firstName":"Melkamu","middleName":"Hunegnaw","lastName":"Asmare","suffix":""},{"id":605082670,"identity":"e90e1e71-4cf9-4869-b7c6-599a408d7c32","order_by":3,"name":"Daniel Cherkos Teka","email":"","orcid":"","institution":"Addis Ababa University College of Health Sciences","correspondingAuthor":false,"prefix":"","firstName":"Daniel","middleName":"Cherkos","lastName":"Teka","suffix":""},{"id":605082671,"identity":"aab75736-1426-4427-b5a6-1747ae3783ea","order_by":4,"name":"Saba Habtu Hagos","email":"","orcid":"","institution":"Addis Ababa University College of Health Sciences","correspondingAuthor":false,"prefix":"","firstName":"Saba","middleName":"Habtu","lastName":"Hagos","suffix":""},{"id":605082672,"identity":"c632aa16-4f96-4edd-b24e-fb8dcfa56cff","order_by":5,"name":"Eilham Elias Jobir","email":"","orcid":"","institution":"Addis Ababa University College of Health Sciences","correspondingAuthor":false,"prefix":"","firstName":"Eilham","middleName":"Elias","lastName":"Jobir","suffix":""},{"id":605082673,"identity":"e0a8fe58-3350-4a42-ba3c-00e11128fc3d","order_by":6,"name":"Abenezer Shiferaw","email":"","orcid":"","institution":"Addis Ababa University College of Health Sciences","correspondingAuthor":false,"prefix":"","firstName":"Abenezer","middleName":"","lastName":"Shiferaw","suffix":""},{"id":605082674,"identity":"1239cca7-59af-4da3-85c8-e609642aa229","order_by":7,"name":"Tsega Woldegiorgis Godebo","email":"","orcid":"","institution":"Mekelle University College of Health Sciences","correspondingAuthor":false,"prefix":"","firstName":"Tsega","middleName":"Woldegiorgis","lastName":"Godebo","suffix":""},{"id":605082675,"identity":"6d2df8c4-8628-448d-bfcf-c2a3ff509ad6","order_by":8,"name":"Akrem Kedir Shemsu","email":"","orcid":"","institution":"Addis Ababa University College of Health Sciences","correspondingAuthor":false,"prefix":"","firstName":"Akrem","middleName":"Kedir","lastName":"Shemsu","suffix":""}],"badges":[],"createdAt":"2026-03-12 06:54:47","currentVersionCode":1,"declarations":{"humanSubjects":true,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":true,"humanSubjectConsent":true,"humanSubjectClinicalTrial":true,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-9101106/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9101106/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":104583121,"identity":"8778e841-482f-4e5e-947f-027772f1cfa8","added_by":"auto","created_at":"2026-03-13 15:18:05","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":133595,"visible":true,"origin":"","legend":"\u003cp\u003eStudy flow diagram showing enrolment, data linkage, scan outcomes, and referral pathway across four urban health centers in Addis Ababa during the 4-week pilot.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-9101106/v1/af81cb23bfba6b5c94d65385.png"},{"id":104583123,"identity":"a7226c1a-1979-4da3-9522-ad2c6f7da5f6","added_by":"auto","created_at":"2026-03-13 15:18:05","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":50881,"visible":true,"origin":"","legend":"\u003cp\u003eSite-wise scan volume and referral yield across the four participating health centers. Scan volume varied considerably by site, with Woreda 7 contributing the highest number of scans and Kebena the lowest.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-9101106/v1/358573d8c870a31aa3f4f359.png"},{"id":104583125,"identity":"7a28a35e-bede-4be7-84da-f8d436c748b9","added_by":"auto","created_at":"2026-03-13 15:18:05","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1397044,"visible":true,"origin":"","legend":"\u003cp\u003eRepresentative handheld obstetric POCUS images from the pilot, captured using Butterfly iQ devices.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePanel A: \u003c/strong\u003eEarly first-trimester intrauterine pregnancy. B-mode image showing a gestational sac with measured gestational sac diameter (GSD2 = 17.26 mm), corresponding to an estimated gestational age of 6 weeks and 6 days. Additional measurements (GSD1 = 19.03 mm) are displayed. Device parameters: TIS 0.01, MI 0.17, OB 1/GYN preset.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePanel B: \u003c/strong\u003eNormal fetal cardiac activity in a second-trimester pregnancy. Doppler spectral tracing (upper panel) demonstrates a regular fetal heart rhythm with a measured heart rate of 156 beats per minute (HR 0.38 s). B-mode image (lower panel) shows the Doppler gate placement. Device parameters: TIS 0.14, MI 0.24, OB 2/3 preset.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePanel C: \u003c/strong\u003eUltrasound showed a bifid uterine cavity with a 17 mm central septum and minimal endometrial thickening, consistent with a subseptate uterus, a congenital anomaly confirmed by a consultant radiologist. Device parameters: TIS 0.01, MI 0.17, OB/GYN preset\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePanel D: \u003c/strong\u003eAmniotic fluid index (AFI) is normal at 16.4 cm in a well-visualized male fetus. Device parameters: TIS 0.01, MI 0.17: OB 2/3 preset\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePanel E\u003c/strong\u003e: Cephalic fetal presentation with back anterior, anterior–fundal placenta, appearing normal. Device parameters : TIS 0.01 , MI 0.17 , OB 2/3\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-9101106/v1/075f077975ae0c1051e52a54.png"},{"id":104781699,"identity":"cbd5b9cd-94d9-47a5-a1fa-626ebd2e3c22","added_by":"auto","created_at":"2026-03-17 07:56:11","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2492853,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9101106/v1/8ad2255c-955a-477c-832e-03273db29a84.pdf"},{"id":104583122,"identity":"bbfde315-7b83-4622-9930-b9e7c25d45de","added_by":"auto","created_at":"2026-03-13 15:18:05","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":206317,"visible":true,"origin":"","legend":"","description":"","filename":"Supplentarymaterials.docx","url":"https://assets-eu.researchsquare.com/files/rs-9101106/v1/d6f1eee59ac749d7686a1417.docx"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eFeasibility and early clinical yield of handheld AI-assisted obstetric point-of-care ultrasound in routine antenatal care: a pilot study in Addis Ababa, Ethiopia\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"Background","content":"\u003cp\u003eUltrasound is now a core component of modern antenatal care. The World Health Organization recommends one ultrasound examination before 24 weeks\u0026rsquo; gestation to improve gestational-age estimation, detect multiple pregnancy and fetal anomalies earlier, reduce post-term induction, and improve the pregnancy experience [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. At the same time, WHO guidance does not support routine fetal Doppler as a standard population-level ANC intervention for all pregnancies [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eDespite that recommendation, access to timely obstetric ultrasound remains uneven across sub-Saharan Africa, where imaging is still often concentrated in hospitals and specialist settings. Recent African implementation studies show that task-shared obstetric POCUS can be delivered in low-resource settings when training, mentorship, supervision, and quality assurance are built into the model. In Kenya, large-scale training and later RE-AIM evaluation showed that scale-up is possible but depends on ongoing mentorship, maintenance, and implementation support [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. In Malawi, midwife-delivered routine ANC POCUS was highly acceptable and achieved acceptable image quality in most reviewed scans [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eEthiopian evidence already supports the value of decentralized portable obstetric ultrasound. A retrospective study from semi-urban health centers reported clinically important abnormal findings in 12.7% of scanned pregnancies and referral of 98.4% of those abnormal cases [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e], while a later pre-post study found that institutionalizing limited obstetric ultrasound was associated with improved ANC, skilled delivery, and postnatal service utilization [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. However, published Ethiopian work has provided less detail on digitally linked image archiving, specialist over-read, and structured quality assurance within routine urban government primary-care workflows.\u003c/p\u003e \u003cp\u003eAgainst this background, our pilot evaluated the structured integration of handheld obstetric POCUS into routine ANC across four urban government health centers in Addis Ababa. The purpose was not to replace formal radiology, but to assess whether a focused handheld ultrasound package could be embedded into primary-care ANC workflows, generate usable focused findings, support referral decisions, and operate within a digitally linked quality-assurance pathway. We aimed to evaluate the feasibility of workflow integration, the focused obstetric findings patterns, and the referral implications of handheld obstetric POCUS embedded in routine ANC at four urban primary health centers in Addis Ababa.\u003c/p\u003e"},{"header":"Methods","content":"\u003ch2\u003eStudy design and setting\u003c/h2\u003e\n\u003cp\u003eThis study was designed as a prospective, multi-center feasibility and implementation pilot to evaluate the integration of handheld AI-enabled obstetric point-of-care ultrasound (POCUS) into routine antenatal care (ANC) services at the primary health care level in Addis Ababa, Ethiopia. Butterfly Network is approved in Ethiopia but used in a few facilities, with limited exposure outside major teaching hospitals. The pilot was conducted over a fixed 4-week implementation period in four urban government health centers: Abuare, Signal-Woreda 7, Kebena, and Janmeda, following submission and institutional review of an integrated obstetric POCUS proposal and in coordination with local clinical radiology leadership and relevant health authorities. Rather than establishing a stand-alone research clinic, the intervention was embedded within existing ANC workflows to assess real-world operational feasibility, uptake, and clinical usefulness at the primary-care level. All the Butterfly OB POCUS was reported on the mother\u0026rsquo;s record indicating effective and comparative outcome.\u003c/p\u003e\n\u003cp\u003eThe pilot was conceived as an implementation-oriented service integration study rather than an efficacy trial. The primary purpose was to determine whether a structured package of focused obstetric ultrasound, supported by standardized documentation, digital image storage, and specialist oversight, could be operationalized within routine ANC consultations in urban health centers. The study also sought to characterize the pattern of focused obstetric findings generated in this setting and to document whether scan findings led to continued management at the health center or referral to a higher level of care.\u003c/p\u003e\n\u003ch2\u003eParticipants and eligibility\u003c/h2\u003e\n\u003cp\u003eAll consecutive pregnant women presenting for routine ANC at participating health centers during the study period were considered for inclusion. Eligibility was intentionally broad because the pilot aimed to evaluate service integration under routine practice conditions rather than under highly restricted trial criteria. Women attending ANC who agreed to undergo focused obstetric POCUS as part of the visit were enrolled, allowing the study to capture the case mix typically encountered in primary-care ANC, including both early pregnancy viability/dating encounters and later gestation follow-up visits. Participation was voluntary, and women who declined an ultrasound continued to receive standard ANC without affecting their routine care.\u003c/p\u003e\n\u003cp\u003eAt enrollment, standardized baseline information was collected using structured study tools. Variables recorded at the point of care included maternal age, gravidity, parity, gestational age by last menstrual period when available, prior obstetric history, ANC follow-up status, maternal education level, residence type, and predefined high-risk clinical indicators such as vaginal bleeding, abdominal pain, reduced fetal movement, or hypertensive symptoms. Immediate clinical disposition after scanning, including continued management at the primary-care facility or referral to a higher-level facility, was also documented. The analytic dataset ultimately included 101 enrolled women, each linked to a unique study identifier.\u003c/p\u003e\n\u003ch2\u003eIntervention and training\u003c/h2\u003e\n\u003cp\u003eThe intervention consisted of a structured introduction of handheld obstetric POCUS into routine ANC using Butterfly ultrasound devices connected to secure study mobile phones, standardized focused scan forms, defined data-entry procedures, and consultant Radiologist specialist-supported oversight. Service delivery was based on a task-shared model involving five junior general practitioners together with site-based midwives, with two midwives participating at each health center. This model was selected to reflect the existing primary maternal health workforce and to test whether focused obstetric ultrasound could be incorporated into routine care without creating a parallel imaging service. Before implementation, participating providers underwent institutional orientation and focused practical training in core obstetric POCUS applications relevant to primary care. Training covered four linked domains: image acquisition, focused interpretation within scope, standardized documentation on the obstetric POCUS form, and escalation of uncertain or abnormal findings through the specialist review pathway. Three Butterfly handheld ultrasound devices were deployed across the four participating sites and connected to secure mobile phones with administrator-controlled access.\u003c/p\u003e\n\u003ch2\u003eFocused scan package\u003c/h2\u003e\n\u003cp\u003eThe ultrasound protocol used in this pilot was a focused obstetric POCUS package rather than a comprehensive formal obstetric sonography examination. The structured acquisition form captured the following core domains at the time of scanning: fetal cardiac activity, gestational age by ultrasound, fetal presentation, placental location, qualitative assessment of amniotic fluid, gross fetal abnormality screening, and Doppler assessment. In clinically relevant early pregnancy cases, the focused examination also supported confirmation of intrauterine pregnancy and assessment of viability concerns.\u003c/p\u003e\n\u003cp\u003eThe focused package was designed to support clinically actionable bedside decision-making in ANC rather than to replace formal radiology or full fetal anomaly scanning. Accordingly, fetal presentation, placental location, and amniotic fluid were recorded as focused screening variables, while \u0026ldquo;gross abnormality suspected\u0026rdquo; functioned as a trigger for further review rather than as a definitive diagnostic classification.\u003c/p\u003e\n\u003ch2\u003eData capture and quality assurance\u003c/h2\u003e\n\u003cp\u003eData capture used a linked clinical-imaging workflow designed to preserve standardization while allowing case-level review. At the point of care, maternal characteristics and focused ultrasound findings were recorded using a standardized obstetric POCUS data collection tool. Ultrasound images were then stored in the secure Butterfly cloud-based application using administrator-controlled accounts, while corresponding clinical and demographic variables were entered into REDCap using unique study identifiers. This architecture enabled de-identified linkage between scan findings and stored images without exporting direct patient identifiers into the analytic dataset. The REDCap dataset used study IDs ranging from 1 to 101. The Butterfly iQ3 ultrasound system includes automated iQ Fan and iQ Slice tools, multiple imaging modes, and over 20 anatomical presets, enabling comprehensive obstetric scanning, looking at details, effective umbilical doppler assessment, high image resolution, and facilitating easy training of research assistants during the pilot.\u003c/p\u003e\n\u003cp\u003eTo improve data integrity, REDCap fields were structured in predefined columns and incorporated validation rules and range checks. Access to the Butterfly platform and REDCap database was restricted to authorized study personnel, HIPAA-regulated, and only de-identified data were exported for analysis. The pilot also incorporated a quality-assurance pathway through specialist-supported oversight, with scanning performed under radiologist supervision, creating opportunities for teleconsultation and expert confirmation of findings.\u003c/p\u003e\n\u003ch2\u003eOutcomes\u003c/h2\u003e\n\u003cp\u003eThe study outcomes were specified to capture both the implementation feasibility of integrating handheld obstetric POCUS into routine antenatal care and the early clinical yield of the focused scan package in a primary health care setting.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePrimary outcomes\u003c/strong\u003e included: (i) feasibility of routine ANC integration, defined as successful incorporation of focused obstetric POCUS into the existing ANC workflow at each of the four participating health centers during the pilot period, without the need to establish a parallel imaging service; (ii) completeness of focused scan fields, defined by the proportion of scans for which the core structured variables were recorded in REDCap; (iii) proportion of scans classified as interpretable versus indeterminate, using the structured recording options for variables marked as \u0026ldquo;unable to determine,\u0026rdquo; \u0026ldquo;uncertain,\u0026rdquo; or \u0026ldquo;not diagnostic\u0026rdquo;; (iv) focused finding spectrum, defined as the distribution of principal ultrasound findings identified during the pilot; and (v) referrals generated, defined as the number and proportion of women whose scan findings resulted in referral from the health center to a higher-level facility.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSecondary outcomes\u003c/strong\u003e included the proportion of women receiving their first-ever ultrasound in the current pregnancy, the overall abnormality detection rate, the radiologist over-read concordance rate, final expert-reviewed abnormality categories, and site-level variation in scan volume, completeness, finding patterns, indeterminate scan fields, and referral frequency across the participating health centers. Secondary outcomes also included accuracy and completeness of AI-assisted fetal biometric measurements used for gestational age estimation and variation in clinical detection patterns across trimesters.\u003c/p\u003e\n\u003ch2\u003eStatistical analysis\u003c/h2\u003e\n\u003cp\u003eBecause this was a feasibility pilot, analyses were primarily descriptive. Continuous variables were summarized using median (IQR) and mean (SD), and categorical variables using counts and percentages with non-missing denominators. Exact 95% confidence intervals were calculated for key proportions. No formal power calculation was performed, and any site-level comparisons were considered exploratory. Butterfly IQ 3 ultrasound was AI-powered, and most biometric calculations were automatic.\u003c/p\u003e\n\u003ch2\u003eEthics\u003c/h2\u003e\n\u003cp\u003eEthical approval was obtained from the relevant institutional body. Written informed consent was obtained from all participants before enrollment. Participation was voluntary, and women retained the right to decline an ultrasound without any effect on their routine ANC care. Ultrasound images were stored on the secure Butterfly cloud platform, and clinical data were entered into password-protected REDCap databases using de-identified study IDs. Access was restricted to authorized study personnel, and no directly identifiable patient information was included in analytic exports.\u003c/p\u003e"},{"header":"Results","content":"\u003ch2\u003eCohort and site distribution\u003c/h2\u003e\n\u003cp\u003eA total of 101 pregnant women underwent focused obstetric POCUS during the 4-week pilot across four government health centers in Addis Ababa. Site distribution was Signal-Woreda 7, 37/101 (36.6%); Janmeda, 33/101 (32.7%); Abuare, 22/101 (21.8%); and Kebena, 9/101 (8.9%). Mean maternal age was 28.3 years (SD 4.4; range 18\u0026ndash;39). Median gravidity was 2 (IQR 1\u0026ndash;3), and median parity was 1 (IQR 0\u0026ndash;2). Among participants with residence data, 98/101 (99.0%) were urban residents. The highest education level was primary in 38/101 (38.0%), secondary in 30/100 (30.0%), and college in 31/100 (31.0%), with 1/100 (1.0%) reporting no formal education. Full participant characteristics are presented in Table 1.\u003c/p\u003e\n\u003ch2\u003eGestational-age and viability profile\u003c/h2\u003e\n\u003cp\u003eUltrasound-estimated gestational age was recorded for 100/101 women (1 missing). The median gestational age was 30 weeks (IQR 16.8\u0026ndash;33.0), with a mean of 25.7 weeks (SD 10.3; range 0\u0026ndash;39). By trimester distribution, the first trimester accounted for approximately 19/101, the second trimester for approximately 24/101, and the third trimester for approximately 58/101 scans. Fetal cardiac activity was documented as present in 92/99 scans (92.9%), absent in 4/99 (4.0%) fetal demises, and indeterminate (unable to determine) in 3/99 (3.0%), mostly in early 1st trimester pregnancies, with 2 values missing.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFocused obstetric findings\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAcross all scans, fetal presentation was recorded as cephalic in 65/101 (64.4%), breech in 5/101 (5.0%), transverse in 17/101 (16.8%), and unable to determine in 14/101 (13.9%). Placental location was anterior in 54/100 (54.0%), posterior in 22/100 (22.0%), fundal in 16/100 (16.0%), low-lying in 1/100 (1.0%), and uncertain in 7/100 (7.0%), with 1 value missing. Uncertain placental location was predominantly reported in early first-trimester pregnancies. Qualitative amniotic fluid was normal in 95/100 (95.0%), low in 1/100 (1.0%), and uncertain in 4/100 (4.0%), with 1 value missing and no cases of high fluid. One congenital uterine abnormality subseptate uterus, was detected (Fig 3, Panel 3 ). Complete focused obstetric POCUS findings are detailed in Table 2. In one case, a mother at 37 weeks + 3 days gestation was initially diagnosed with anhydramnios and recommended for emergency cesarean section by MFM obstetrician scan. Re-evaluation using Butterfly POCUS, in consultation with a radiologist, identified a posterior amniotic fluid pocket measuring 11 cm with a reassuring biophysical profile (BPP). The mother was subsequently followed and later had a safe vaginal delivery.\u003c/p\u003e\n\u003ch2\u003eReferral and non-normal findings\u003c/h2\u003e\n\u003cp\u003eA binary normal-scan indicator classified 93/101 scans (92.1%) as normal and 8/101 (7.9%) as non-normal. Seven of 99 women (7.1%) were referred following the scan encounter, with 2 values missing for referral status. Site-level implementation outcomes are presented in Table 3.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 1.\u0026nbsp;\u003c/strong\u003eParticipant characteristics (N = 101)\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"624\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 233px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCharacteristic\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 195px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eValue\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 195px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eN (non-missing)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 233px;\"\u003e\n \u003cp\u003eAge (years), mean \u0026plusmn; SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e28.3 \u0026plusmn; 4.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e101\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 233px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Range\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e18\u0026ndash;39\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 233px;\"\u003e\n \u003cp\u003eGravidity, median (IQR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e2 (1\u0026ndash;3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e101\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 233px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Range\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e1\u0026ndash;6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 233px;\"\u003e\n \u003cp\u003eParity, median (IQR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e1 (0\u0026ndash;2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e101\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 233px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Range\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e0\u0026ndash;4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 233px;\"\u003e\n \u003cp\u003eHighest education, n/N (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e100\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 233px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;None\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e1/100 (1.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 233px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Primary\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e38/100 (38.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 233px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Secondary\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e30/100 (30.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 233px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;College\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e31/100 (31.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 233px;\"\u003e\n \u003cp\u003eResidence, n/N (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e99\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 233px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Urban\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e98/99 (99.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 233px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Peri-urban\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e1/99 (1.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 233px;\"\u003e\n \u003cp\u003eHealth center site, n/N (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e101\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 233px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Woreda 7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e37/101 (36.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 233px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Janmeda\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e33/101 (32.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 233px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Abuare\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e22/101 (21.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 233px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Kebena\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e9/101 (8.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 195px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cem\u003eSD, standard deviation; IQR, interquartile range.\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2.\u0026nbsp;\u003c/strong\u003eFocused obstetric POCUS findings (N = 101)\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"624\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 253px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eFinding\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 185px;\"\u003e\n \u003cp\u003e\u003cstrong\u003en/N (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 185px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMissing\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003eFetal cardiac activity\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e2 (2.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Present\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e92/99 (92.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Absent\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e4/99 (4.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Unable to determine\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e3/99 (3.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003eGestational age (weeks)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e1 (1.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Mean \u0026plusmn; SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e25.7 \u0026plusmn; 10.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Median (IQR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e30 (16.8\u0026ndash;33.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003eCalculated trimester\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e0 (0.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;First (\u0026le;13 weeks)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e19/101 (18.8%)\u0026dagger;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Second (14\u0026ndash;27 weeks)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e24/101 (23.8%)\u0026dagger;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Third (\u0026ge;28 weeks)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e58/101 (57.4%)\u0026dagger;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003eFetal presentation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e0 (0.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Cephalic\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e69/101 (68.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Breech\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e5/101 (5.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Transverse\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e17/101 (16.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Unable to determine\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e10/101 (9.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003ePlacental location\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e1 (1.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Anterior\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e54/100 (54.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Posterior\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e22/100 (22.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Fundal\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e16/100 (16.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Low-lying\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e1/100 (1.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Uncertain\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e7/100 (7.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003eAmniotic fluid (qualitative)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e1 (1.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Normal\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e95/100 (95.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Low\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e1/100 (1.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;High\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e0/100 (0.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Uncertain\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e4/100 (4.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003eNormal scan indicator\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e0 (0.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Normal (1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e93/101 (92.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Non-normal (0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e8/101 (7.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003eReferral made\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e2 (2.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e7/99 (7.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 253px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e92/99 (92.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 185px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cem\u003e\u0026dagger; Trimester proportions derived from calculated trimester field (mean 2.44, SD 0.74). Percentages are approximate and may vary slightly based on gestational age coding.\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 3.\u0026nbsp;\u003c/strong\u003eSite-level implementation outcomes\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"624\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eIndicator\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSignal-Woreda 7\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eJanmeda\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAbuare\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 125px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eKebena\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eTotal scans, n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e37 (36.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e33 (32.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e22 (21.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e9 (8.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eComplete forms, n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e37 (100)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e33 (100)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e22 (100)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e9 (100)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eIndeterminate presentation, n\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eUncertain placental location, n\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eUncertain amniotic fluid, n\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eAbnormal scans, n\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eReferrals, n\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cem\u003e* NB-Most undetermined fetal presentations were recorded by GPs in the second trimester, when the fetus is mobile and placental location may be less clearly defined. Two cases of oligohydramnios and two fetal demises were categorized as undetermined AFI. Nearly all scans reviewed by the supervising radiologist demonstrated clear biometrics and definitive assessment of placental location, AFI, and fetal presentation.\u003c/em\u003e\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this 4-week pilot, handheld focused obstetric POCUS was integrated into routine ANC across four urban government health centers in Addis Ababa, with 101 women scanned and low missingness across most core variables. Most scans were classified as normal, but a clinically meaningful minority showed absent or indeterminate fetal cardiac activity, non-cephalic presentation, abnormal or uncertain placental or fluid findings, or required referral. These findings support the operational feasibility of embedding a focused handheld ultrasound package in primary-care ANC, while also showing the importance of capturing indeterminate and uncertain results rather than masking them as missing data.\u003c/p\u003e \u003cp\u003eThe findings align with previous Ethiopian studies showing that portable obstetric ultrasound at the health-center level can detect important abnormalities and trigger referrals [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. The novelty of this study lies in integrating handheld POCUS with REDCap data capture, cloud image archiving, and specialist-supported review, addressing prior quality-assurance gaps [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. The broader African implementation literature strengthens this interpretation. In Kenya, large-scale programmatic rollout showed that non-radiologist obstetric POCUS can be scaled when training, supervision, and health-system support are sustained [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. In Malawi, midwife-performed ANC POCUS was highly acceptable and achieved acceptable scan quality in most reviewed scans [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Together, these studies suggest that feasibility in low-resource settings depends less on the handheld device alone than on the surrounding implementation package: training, mentorship, image review, maintenance, consumables, and referral feedback.\u003c/p\u003e \u003cp\u003eAccess to imaging services plays a vital role in the diagnosis and treatment of a range of clinical conditions. However, low- and middle-income countries experience significant shortages in both imaging equipment and trained personnel [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. POCUS has emerged as an important tool for improving access to diagnostic imaging in resource-limited primary care settings [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e], with AI-enabled handheld devices such as the Butterfly iQ offering additional potential for standardized gestational-age estimation, anatomical scans, and decision support at the point of care [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. The 2016 WHO antenatal care guidelines recommend at least one ultrasound before 24 weeks to estimate gestational age, detect fetal anomalies and multiple pregnancies, reduce post-term labor induction, and enhance the pregnancy experience [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. AI-assisted functionality was integrated into the handheld POCUS platform to support fast image acquisition, automated fetal biometric measurements, and streamlined obstetric workflow even for beginners. AI protocols enabled simplified gestational age estimation through blind-sweep scanning and multi-slice imaging, while decision-support tools such as Fast Maternal and Fetal Ultrasound (FAMLI) guided evaluation of key obstetric conditions, including malpresentation, abnormal amniotic fluid, multiple gestation, and fetal viability [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Semi-automated documentation further improved reporting efficiency, supporting diagnostic consistency and safe task-shifting in primary care settings [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe implementation barriers reported elsewhere in Ethiopia and across LMIC settings are also relevant to this pilot. These include staff workload, shortages of trained personnel, service interruptions, consumable supply, maintenance, electricity, connectivity, and access to second-line review [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Recent Ethiopian tele-ultrasound experience suggests that remote specialist support can reduce travel burden and improve case escalation, making the Butterfly-plus-REDCap workflow in this study especially relevant to future scale-up models.\u003c/p\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eLimitations\u003c/h2\u003e \u003cp\u003eThis pilot study has limitations. It was conducted over a short 4-week period in four urban primary care centers with a small sample size (n\u0026thinsp;=\u0026thinsp;101), limiting statistical power and the ability to draw firm conclusions regarding diagnostic accuracy, clinical outcomes, and long-term impact. The absence of a control group or pre\u0026ndash;post comparison further limits causal interpretation. The study was performed in urban facilities with relatively reliable electricity and connectivity, which may not reflect conditions in rural settings. Limited handheld ultrasound devices and compatible mobile phones also constrained scan throughput at some sites.\u003c/p\u003e \u003cp\u003eAlthough scans were supervised by a consultant radiologist and met quality standards, formal inter-observer testing and comparison with comprehensive radiology ultrasound were not performed, limiting assessment of diagnostic accuracy. Operational metrics and patient follow-up were also limited, and effective POCUS implementation requires structured training and supervision for primary healthcare providers. Future studies should include larger and more diverse populations, longer follow-up, and evaluation of diagnostic accuracy, referral completion, and maternal\u0026ndash;neonatal outcomes. While the pilot demonstrates feasibility and early clinical yield, it does not yet establish improvements in outcomes, mortality, or referral appropriateness.\u003c/p\u003e \u003c/div\u003e"},{"header":"Conclusions","content":"\u003cp\u003eHandheld obstetric POCUS was feasibly integrated into routine antenatal care across four urban government health centers in Addis Ababa and identified clinically actionable findings requiring referral in a subset of women. Health center staff, including midwives and frontline health workers, demonstrated strong engagement and expressed a highly positive attitude toward the service, noting that the scans were carefully performed and valuable for patient care, and they supported its incorporation into routine practice.\u003c/p\u003e\n\u003cp\u003eImplemented within Ministry-aligned governance, competency-based training, and consultant radiologist supervision, decentralized ultrasound shows promise for strengthening early detection of high-risk pregnancies at the primary care level. Larger prospective studies with standardized adjudication of abnormality, image-quality metrics, and referral follow-up are needed to confirm diagnostic performance, clinical impact, and scalability.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNo specific funding was received for this study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflicts of interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no conflicts of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eData will be co-owned by the primary investigators and securely stored on the Butterfly Cloud platform and the University of Washington REDCap system. De-identified data will be available upon reasonable ethical request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026rsquo; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCR and MG conceptualized the study. The field work was coordinated by MG. CR, MG, DC, SH, EJ, AS, BW, and AS contributed to data collection. Data analysis and interpretation were conducted by MH, MG, and CR. All authors have read and approved the final document.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors thank all study participants, the volunteering GPs, midwives, and staff of the participating health centers for their support and collaboration. The investigators also acknowledge Riverstone Health for supplying the Butterfly handheld ultrasound devices used in this study, and the University of Washington for providing REDCap access for data management and structuring via the WWAMI Program.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eWorld Health Organization. WHO Recommendations on Antenatal Care for a Positive Pregnancy Experience. Geneva: World Health Organization; 2016. Available from: https://www.who.int/publications/i/item/9789241549912\u003c/li\u003e\n\u003cli\u003eWachira J, Matheka DM, Masheti SA, Githemo GK, Shah S, Haldeman MS, Ramos M, Bergman K. A training program for obstetrics point-of-care ultrasound to 514 rural healthcare providers in Kenya. BMC Med Educ. 2023;23(1). doi:10.1186/s12909-023-04886-x\u003c/li\u003e\n\u003cli\u003eGithemo G, Wanyoro A, Masika J, Onsongo L, Bett S, Githuku S, Gachuiri G, Walker D, Santos N, Ghosh R, Otieno G. Evaluation of large-scale implementation of obstetric point-of-care ultrasound in eight counties in Kenya using RE-AIM framework. BMC Health Serv Res. 2025;25(1):1016. doi:10.1186/s12913-025-13212-8\u003c/li\u003e\n\u003cli\u003ePayesa C, Seyama L, Chimwaza Y, Sindani F, Kanise Y, Bvutula E, Phiri M, Nyangulu P, Gadama L, Kachale F, Gadama G, Mwale M, Yenokyan G, Sripad P, Hyre A, Noguchi LM, Dadabhai S. Feasibility and acceptability of point-of-care ultrasound delivered by midwives during routine antenatal care in Malawi: a prospective implementation science study. BMJ Open. 2025;15(8). doi:10.1136/bmjopen-2025-100515\u003c/li\u003e\n\u003cli\u003eAbawollo HS, Tsegaye ZT, Desta BF, Beshir IA, Mengesha BT, Guteta AA, Heyi AF, Mamo TT, Gebremedhin ZK, Damte HD, Zelealem M, Argaw MD. Contribution of portable obstetric ultrasound service innovation in averting maternal and neonatal morbidities and mortalities at semi-urban health centers of Ethiopia: a retrospective facility-based study. BMC Pregnancy Childbirth. 2022;22(1):368. doi:10.1186/s12884-022-04703-1\u003c/li\u003e\n\u003cli\u003eWoldeyohannes D, Abera M, Gebremedhin S, Tesfaye G. Institutionalization of limited obstetric ultrasound leading to increased antenatal, skilled delivery, and postnatal service utilization in three regions of Ethiopia: a pre-post study. PLoS One. 2023. doi:10.1371/journal.pone.0281626\u003c/li\u003e\n\u003cli\u003eFrija G, Blažić I, Frush DP, Hierath M, Kawooya M, Donoso-Bach L, Brkljačić B. How to improve access to medical imaging in low- and middle-income countries? eClinicalMedicine. 2021;38:101034. doi:10.1016/j.eclinm.2021.101034\u003c/li\u003e\n\u003cli\u003eFleming KA, Horton S, Wilson ML, Atun R, DeStigter K, Flanigan J, et al. The Lancet Commission on diagnostics: transforming access to diagnostics. Lancet. 2021;398(10315):1997\u0026ndash;2050. doi:10.1016/S0140-6736(21)00673-5\u003c/li\u003e\n\u003cli\u003eDella Ripa S, Santos N, Walker D. AI-enabled obstetric point-of-care ultrasound as an emerging technology in low- and middle-income countries: provider and health system perspectives. BMC Pregnancy Childbirth. 2025;25:729. doi:10.1186/s12884-025-07796-6\u003c/li\u003e\n\u003cli\u003eRanger BJ, Bradburn E, Chen Q, Kim M, Noble JA, Papageorghiou AT. Portable ultrasound devices for obstetric care in resource-constrained environments: mapping the landscape. Gates Open Res. 2024;7:133. doi:10.12688/gatesopenres.15088.2\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"RiverStone Health","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Obstetric POCUS, Point-of-care ultrasound, Antenatal care, Ethiopia, Butterfly ultrasound, Primary health care, Implementation, Feasibility","lastPublishedDoi":"10.21203/rs.3.rs-9101106/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9101106/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground: \u003c/strong\u003eThe World Health Organization recommends at least one ultrasound examination before 24 weeks of gestation to improve gestational-age assessment and detection of multiple pregnancy and fetal anomalies. However, access to obstetric ultrasound at the primary-care level remains limited in many low-resource settings, including Ethiopia. This study evaluated the feasibility and early clinical yield of integrating AI-enabled handheld obstetric point-of-care ultrasound (POCUS) into routine antenatal care in four government health centers in Addis Ababa.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods: \u003c/strong\u003eWe conducted a 4-week prospective pilot across four urban health centers (Abuare, Signal-Woreda 7, Kebena, and Janmeda). Pregnant women attending routine antenatal care (ANC) underwent focused obstetric POCUS using Butterfly handheld devices. Trained frontline providers, mainly junior GPs and midwives, performed scans using a standardized acquisition form, while clinical data were entered into REDCap and linked to archived image records on the Butterfly cloud platform. Primary outcomes were the feasibility of workflow integration, completeness of focused scan variables, the spectrum of focused obstetric findings, and referrals generated.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults: \u003c/strong\u003eA total of 101 women were scanned. Mean maternal age was 28.3 years (SD 4.4); median gravidity was 2 (IQR 1–3), and median parity was 1 (IQR 0–2). Ultrasound-estimated gestational age was recorded in 100/101 women (median 30 weeks; IQR 16.8–33.0). Fetal cardiac activity was present in 92/99 (92.9%), absent in 4/99 (4.0%), and indeterminate in 3/99 (3.0%). Fetal presentation was cephalic in 65/101 (64.4%), breech in 5/101 (5.0%), transverse in 17/101 (16.8%), and indeterminate in 14/101 (13.9%). Placental location was low-lying in 1/100 (1.0%), and qualitative amniotic fluid was low in 1/100 (1.0%). Seven of 99 women (7.1%) were referred to a higher-level facility. A binary normal-scan indicator classified 93/101 scans (92.1%) as normal.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions: \u003c/strong\u003eHandheld focused obstetric POCUS was operationally feasible within routine urban ANC and identified a clinically relevant minority of women requiring referral or further review. Larger studies with standardized expert over-read, referral follow-up, and explicit image-quality metrics are needed before stronger claims about referral optimization or broader clinical impact can be made.\u003c/p\u003e","manuscriptTitle":"Feasibility and early clinical yield of handheld AI-assisted obstetric point-of-care ultrasound in routine antenatal care: a pilot study in Addis Ababa, Ethiopia","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-13 15:18:00","doi":"10.21203/rs.3.rs-9101106/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"ea894265-ff2a-4d18-b84c-a4a1bb2dab91","owner":[],"postedDate":"March 13th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":64393768,"name":"Maternal \u0026 Fetal Medicine"},{"id":64393769,"name":"Biomedical Engineering"},{"id":64393770,"name":"Artificial Intelligence and Machine Learning"}],"tags":[],"updatedAt":"2026-04-23T19:24:23+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-13 15:18:00","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9101106","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9101106","identity":"rs-9101106","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}