Publications

The design of medical systems for remote, resource-limited environments faces persistent challenges due to poor interoperability, lack of offline support, and dependency on costly infrastructure. Many existing digital health solutions neglect these constraints, limiting their effectiveness for frontline health workers in underserved regions. This paper presents a portable, edge-enabled Electronic Health Record platform optimized for offline-first operation, secure patient data management, and modular diagnostic integration. Running on small-form factor embedded devices, it provides AES-256 encrypted local storage with optional cloud synchronization for interoperability. As a use case, we integrated a non-invasive anemia screening module leveraging fingernail pallor analysis. Trained on 250 patient cases (27% anemia prevalence) with KDE-balanced data, the Random Forest model achieved a test RMSE of 1.969 g/dL and MAE of 1.490 g/dL. A severity-based model reached 79.2% sensitivity. To optimize performance, a YOLOv8n-based nail bed detector was quantized to INT8, reducing inference latency from 46.96 ms to 21.50 ms while maintaining mAP@0.5 at 0.995. The system emphasizes low-cost deployment, modularity, and data privacy compliance (HIPAA/GDPR), addressing critical barriers to digital health adoption in disconnected settings. Our work demonstrates a scalable approach to enhance portable health information systems and support frontline healthcare in underserved regions.

Rapid diagnosis of infectious diseases, such as monkeypox, is essential for effective containment and treatment, particularly in resource-constrained settings. This study presents an AI-driven diagnos- tic tool optimized for deployment on the NVIDIA Jetson Orin Nano. Several pre-trained architectures were evaluated, with MobileNetV2 and DenseNet121 achieving the best F1-scores of 91.87% and 86.70%, respec- tively, on the Monkeypox Skin Lesion Dataset (MSLD and MSLD v2.0). TensorRT was used for model optimization, leveraging FP32, FP16, and INT8 precision formats to accelerate inference while reducing model size and power consumption. Results show up to 2.52x speedup and improved energy efficiency with minimal accuracy loss. The system was deployed with a Wi-Fi Access Point (AP) hotspot and a web-based interface, en- abling users to upload and analyze images directly through connected devices such as mobile phones. These advancements position the tool as a scalable, efficient, and low-power diagnostic solution for deployment in underserved healthcare settings.

Skin cancer is one of the most common and preventable cancers, yet early detection remains challenging, especially in resource-limited settings with scarce specialized healthcare. This study develops an AI-driven diagnostic tool optimized for embedded systems to ad- dress this gap. Using transfer learning with MobileNetV2, the model was trained for multi-class classification of skin lesions and optimized with TensorRT for deployment on the NVIDIA Jetson Orin Nano. Evalua- tions focused on model size, inference speed, throughput, and power con- sumption, balancing performance with efficiency. The optimized model maintained an F1-score of 65% while significantly reducing model size and energy consumption. Despite not achieving state-of-the-art accuracy, this research prioritizes real-world feasibility, demonstrating A's poten- tial for accessible diagnostics in low-resource environments. The meth- ods presented extend beyond skin cancer detection, with applications in other medical and autonomous systems, highlighting the broader impact of AI-driven solutions on global healthcare accessibility.

Anemia is a widespread global health issue, particularly among young children in low-resource settings. Traditional methods for anemia detection often require expensive equipment and expert knowledge, cre- ating barriers to early and accurate diagnosis. To address these chal- lenges, we explore the use of deep learning models for detecting anemia through conjunctival pallor, focusing on the CP-AnemiC dataset, which includes 710 images from children aged 6-59 months. The dataset is annotated with hemoglobin levels, gender, age, and other demographic data, enabling the development of machine learning models for accurate anemia detection. We use the MobileNet architecture as a backbone, known for its efficiency in mobile and embedded vision applications, and fine-tune our model end-to-end using data augmentation techniques and a cross-validation strategy. Our model implementation achieved an accu- racy of 0.9313, a precision of 0.9374, and an F1 score of 0.9773, demon- strating strong performance on the dataset. To optimize the model for deployment on edge devices, we performed post-training quantization, evaluating the impact of different bit-widths (FP32, FP16, INT8, and INT4) on model performance. Preliminary results suggest that while FP16 quantization maintains high accuracy (0.9250), precision (0.9370) and F1 score (0.9377), more aggressive quantization (INT8 and INT4) leads to significant performance degradation. Overall, our study supports further exploration of quantization schemes and hardware optimizations to assess trade-offs between model size, inference time, and diagnostic accuracy in mobile healthcare applications.