DefinePK hosts the largest index of Pakistani journals, research articles, news headlines, and videos. It also offers chapter-level book search.
Title: AI-POWERED ULTRASOUND INTERPRETATION FOR EARLY DETECTION OF KIDNEY ABNORMALITIES IN RURAL PATIENTS
Authors: Muhammad Awais, Asma Rehman, Rabia Shahzad , Ramsha Zafar , Rabia Khattak , Nageena Ghafoor , Maham Salman , Hafsa Saleem
Journal: Insights-Journal of Health and Rehabilitation
| Category | From | To |
|---|---|---|
| Y | 2024-10-01 | 2025-12-31 |
Publisher: Health And Research Insights (SMC-Private) Limited
Country: Pakistan
Year: 2025
Volume: 3
Issue: 5 (Health and Rehabilitation)
Language: en
DOI: 10.71000/f0jswf03
Keywords: ARTIFICIAL INTELLIGENCETelemedicineDiagnostic imagingRenal ultrasonographyRural HealthKidney DiseasesNephrologyScreening
Background: Chronic kidney disease (CKD) often remains undiagnosed in its early stages, particularly in rural and underserved populations lacking access to nephrologists and diagnostic imaging. Ultrasonography is a first-line tool for kidney evaluation but is limited by operator dependence. Recent advancements in artificial intelligence (AI) have introduced opportunities for automated, standardized interpretation of ultrasound images, potentially improving early detection in low-resource settings.
Objective: To evaluate the effectiveness of AI-assisted ultrasound interpretation in detecting early renal abnormalities among underserved rural populations in Pakistan.
Methods: A cross-sectional study was conducted from January to June 2025 at private rural healthcare centers in Lahore, Faisalabad, and Multan. A total of 200 adult patients underwent renal ultrasonography using portable devices equipped with AI diagnostic software. Inclusion criteria were adults ≥18 years without prior CKD diagnosis. AI findings were compared with interpretations by experienced radiologists. Outcome measures included sensitivity, specificity, accuracy, and inter-observer agreement. Statistical analysis was performed using SPSS v26, and ROC curves were generated to evaluate diagnostic performance.
Results: AI-assisted ultrasound demonstrated a sensitivity of 92.3%, specificity of 89.1%, and accuracy of 90.7% in detecting early renal abnormalities, outperforming radiologist interpretation in certain lesion categories. The AI model showed particularly high performance in identifying increased cortical echogenicity and early hydronephrosis. Agreement between AI and radiologist findings reached 91%, with minimal discordance. The area under the ROC curve for AI was 94.2%, indicating excellent diagnostic capability.
Conclusion: AI-assisted ultrasound interpretation significantly improves early detection of renal abnormalities in rural populations, supporting its integration into primary care to reduce diagnostic disparities and enhance kidney health outcomes.
Loading PDF...
Loading Statistics...