Tikur Anbessa Hospital

ABSTRACTAcute pancreatitis can present with ST‐segment elevation and marked troponin elevation, mimicking acute coronary syndrome. In resource‐limited settings where coronary angiography, cardiac magnetic resonance imaging, and serial high‐sensitivity troponin testing are unavailable, distinguishing acute myocardial injury associated with pancreatitis from Type 1 myocardial infarction can be challenging.

We propose a weakly supervised framework for grounding multiple clinical captions to disease regions in medical images using image level labels without pixel level localization annotations. Unlike approaches that treat all textual information equally, the proposed framework models the hierarchical structure of radiology reports by learning the relative importance of individual captions. Building on the Language meets Vision Transformer (LViT), the framework incorporates three main components: (1) multi-caption encoding using BioClinicalBERT, (2) an attention-based caption importance module trained end-to-end under weak supervision, and (3) a weakly supervised grounding mechanism for disease-region localization. The proposed method achieved a caption importance correlation of 0.9552 with clinician-assigned importance rankings and demonstrated improved attention focus compared with the reported baselines (0.5011 vs. 0.4123, p < 0.01). Qualitative results further suggest that higher-weighted captions contribute more strongly to disease region localization, whereas lower-weighted captions provide supporting clinical context. The framework achieved lesion identification accuracy of 0.9811 under weak supervision. Domain experts evaluated the framework using caption importance correlation, attention focus, localization intersection over union, and lesion identification accuracy. Overall, the findings suggest that incorporating the hierarchical structure of clinical reports may improve the alignment between medical images and their associated textual descriptions and may help clinicians more easily interpret medical images of affected regions. The code developed for this study has been made publicly available on GitHub.

Abstract Significant global disparities in breast cancer outcomes persist, with high mortality rates in low- and middle-income countries driven by late diagnosis and limited access to comprehensive breast cancer care. This study evaluates treatment initiation and one-year outcomes in a cohort of young women and men at a tertiary cancer centre in Ethiopia to identify barriers to care and areas for improvement. This was a prospective cohort study conducted at the Tikur Anbessa Specialized Hospital (TASH) in Addis Ababa. Study participants (females aged 18–39, males aged ≥ 18) with invasive breast cancer were followed for one year to assess initiation of recommended cancer treatment and clinical outcome. One-hundred study participants were included. Among participants with stage I-III breast cancer ( n = 76), only 20% initiated all recommended multimodal oncological treatment. While rates for surgery (90.5%) and initiation of systemic therapy (chemotherapy 86.1%, endocrine therapy 85.5%) were high, 37.8% experienced delayed initiation of chemotherapy (>10 weeks post-surgery). Notably, only 16.4% received adjuvant radiotherapy. At one-year follow-up, disease recurrence rates were high: 29.9% ( n = 20/67, stages I-III). Among stage IV patients with available data, the one-year mortality rate was high: 58.3% ( n = 7/12). The uptake of surgery and systemic treatment was high in the study population, although there were significant delays and limited access to radiotherapy. The high recurrence and mortality rates seen in this breast cancer cohort treated at TASH underscore the urgent need for improved access to timely, comprehensive breast cancer care to improve survival.

ABSTRACT Background The double burden of malnutrition, combining overweight or obesity with anemia is vital in nutritional transitions. Evidence regarding its national distribution and prediction using routine survey data in Ethiopia remains limited. Objective To estimate the national prevalence of individual-level double burden of malnutrition among non-pregnant Ethiopian women aged 15-49 years and evaluate machine-learning models using data from the 2024/25 Ethiopian Demographic and Health Survey. Methods We analyzed 14250 non-pregnant women. Double burden of malnutrition was defined as a body mass index equal to or greater than 25.0 kilograms per square meter with an altitude- and smoking-adjusted hemoglobin concentration below 12.0 grams per deciliter. Data were split into an 80% training set and a 20% holdout test set. Six algorithms were evaluated using cross-validation, while SHapley Additive exPlanations described feature contributions. Results The weighted national double burden of malnutrition prevalence was 8.4% (95% confidence interval 7.9%–8.9%), higher in urban residents (20.0% versus 4.0%) and the richest wealth quintile (26.7% versus 2.7%). Extreme Gradient Boosting achieved an area under the receiver operating characteristic curve of 0.892 (95% confidence interval 0.875–0.909) and an area under the precision-recall curve of 0.764 outperforming logistic regression (area under the receiver operating characteristic curve of 0.781). Extreme Gradient Boosting sensitivity was 82.1%, specificity was 81.4% and the Brier score was 0.068. Household wealth, residence and respondent age were top predictors. Conclusions The double burden of malnutrition affected one in twelve non-pregnant Ethiopian women unevenly across groups. Routine characteristics effectively discriminated double burden of malnutrition cases though external validation is needed prior to clinical use.

BackgroundThere is a pressing need for a sustainable, longitudinal point-of-care ultrasound (POCUS) education model that is contextually relevant, locally led, and systematically evaluated in East Africa. The Toronto Addis Ababa Academic Collaboration in Emergency Medicine (TAAAC-EM), a long-standing partnership between the University of Toronto and Addis Ababa University, developed the East African POCUS (E-POCUS) Training Program Pilot. This was a one-year, longitudinal, hybrid advanced training program tailored to local needs and interests.MethodsWe used reflexive thematic analysis to conduct a qualitative program evaluation of the E-POCUS training program pilot. E-POCUS participants, staff involved with the execution of the training program, and those participants who had to leave the program were all invited to participate in a semi-structured virtual interview, exploring their perceived impact and opinion of the E-POCUS program. Interview transcripts were analyzed using Braun and Clarke’s reflexive thematic analysis framework.ResultsA total of 13 interviews were conducted, representing participants in the E-POCUS program (n = 2), former participants of the E-POCUS program (n = 3), and staff involved in the teaching of the program (n = 8). The analysis highlighted three main themes: (1) time as a limiting factor, (2) expectations versus reality in program delivery, and (3) motivations for participation and the role of accreditation.ConclusionThis study highlighted the value of flexible, longitudinal, and hybrid curriculums in POCUS education. The E-POCUS pilot also illustrated that the hybrid model of education allowed participants to prioritize hands-on skill development teaching during in-person months. Advancing POCUS education in resource-limited settings is critical to improving patient care, and it is important to evaluate efforts at increasing access and education on the topics that are most relevant to their patient population.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s12245-026-01391-y.

Background: Emergency Department provides timely intervention for patients with life-threatening emergency conditions. This is possible through the involvement of families or caregivers during clinical evaluation and management. However, there are scenarios where patients present to the emergency department solely by themselves or by other agents, such as police or passersby, where their presence during the patient's emergency stay is impractical. Understanding the clinical characteristics and outcomes of patients managed without attendants is important for optimizing emergency care delivery and informing health care planning. This study assesses the prevalence, clinical profile, and outcome of patients without attendants in selected governmental hospitals of Addis Ababa. Methods: A prospective cross-sectional study was conducted from August 1 to November 1, 2025, in selected governmental hospitals. After obtaining ethical clearance, data were collected by trained personnel from patients using a data extraction tool and a questionnaire from patient documents/EMR. Data were cleaned and exported to SPSS version 26. Descriptive statistics, including frequencies and percentages, were generated for the prevalence, clinical profile, and outcomes of patients without an attendant in emergency department admissions. Considering a p-value <0.05 with a 95% confidence interval, a logistic regression model was conducted to assess the strength of association between major outcomes and the clinical profile of patients without attendants. Results: 137 (79.2%) participants were male. The median age was 35 years (IQR: 25–46). The prevalence of patients without attendants was 2.57%. The median length of stay was 11.2 days (IQR: 5.6–19.7). 34 (19.6%) patients died. Male sex and high-acuity triage category (red and orange) were significantly associated with patient mortality. Conclusion: The length of stay and mortality rate among patients without attendants are higher than standard ED benchmarks. Mortality is significantly associated with male sex and high-acuity triage category. Keywords: Patient without attendant; prevalence; length of stay; mortality; Addis Ababa.

Africa faces a growing crisis of end-stage organ failure driven by the rising burden of non-communicable diseases and fragile health systems. Behind these figures are preventable deaths, disrupted families, and widening inequities in access to lifesaving care, an inequity that organ transplantation could address. However, access to transplantation remains extremely limited across the continent. According to the 2024 Global Observatory on Donation and Transplantation (GODT) report, only 544 kidney transplants were performed in Africa, representing approximately 0.5% of global transplant activity. These procedures occurred in 36 transplant centers, corresponding to roughly one center per 13.2 million inhabitants, illustrating the structural gap in transplantation services. This work synthesizes a continent-wide expert initiative. In May 2025, the first African Forum on Organ Donation and Transplantation was held in Nouakchott, Mauritania, bringing together more than 300 experts from 12 countries. A Delphi consensus process was conducted to develop a unified framework addressing transplantation challenges in Africa. The Delphi process led to the adoption of the Nouakchott Declaration, a strategic framework aimed at accelerating transplantation system development, reducing preventable mortality, and promoting ethical organ donation. The Declaration outlines eight core principles addressing governance, public awareness, financing, and regional cooperation. It integrates recommendations from the World Health Organization and the Declaration of Istanbul, while adapting them to Africa-specific challenges such as workforce shortages and limited resources. The Nouakchott Declaration provides a structured and context-sensitive roadmap to strengthen organ donation and transplantation systems across Africa. Its implementation could reduce inequities in access to transplantation and improve survival outcomes, while promoting ethical and sustainable healthcare development on the continent.

Abstract Purpose To develop and validate a mental health personal recovery measure for people with mental health conditions (MHCs) in Ethiopia that is grounded in their recovery journey. Methods The study stages were: reviewing existing measures; qualitative exploration of the concept in Ethiopia (n = 25 interviews); expert workshops (n = 3 involving 28 experts); cognitive interviewing for semantic and content validity (n = 20); pilot study for item reduction and exploratory factor analysis (n = 213) and parallel analysis to determine the dimensional structure (n = 213); and assessment of construct (confirmatory factor analysis) and construct validity in an independent sample (n = 401), and test-retest reliability in a sub-sample (n = 48). Construct validity was assessed against symptom severity and disability. Results Participants gave greater emphasis to social recovery than self-actualisation or empowerment; hence, no existing measures had face validity. We piloted 26 items from existing measures and 19 newly developed items. The draft measure was unidimensional, explaining 71% of variance, and had item-test correlations > 0.58. Item reduction resulted in 20-item scale, ReConnect (Recovery and Connection Scale). In the validation study, ReConnect was unidimensional with acceptable goodness of fit indices and test-retest reliability and correlated significantly with symptom severity and functioning. Conclusion ReConnect is a valid and reliable measure of personal recovery in Ethiopia. It has potential applicability to other similar settings where the community has a greater role in the recovery journey.