Varadraj Prabhu Gurupur

With the rapid advancements in artificial intelligence (AI) and machine learning (ML), the role of data engineering has become increasingly critical due to the growing demands for high-quality, large-scale, and well-structured datasets required to train reliable predictive models. Healthcare is one of the most data-intensive industries and has demonstrated strong potential for AI-driven automation in clinical decision support, diagnostics, and operational efficiency. However, healthcare data is often fragmented across multiple systems, inconsistently formatted, and constrained by privacy and regulatory requirements, creating significant barriers to scalable AI adoption. In this review, we examine recent research on healthcare data engineering and AI applications, focusing on how data pipelines, interoperability, and governance frameworks support or limit real-world deployment. This review examined 68 peer-reviewed studies published between 2018 and 2026 across multiple clinical domains, including oncology, cardiovascular disease, infectious disease, neurological disorders, medical imaging, and algorithmic frameworks for explainability and fairness. The reviewed literature shows that while AI models achieve promising performance across these domains, the lack of standardized data architectures and interoperable infrastructure remains a primary bottleneck. The purpose of this study is to highlight key challenges and emerging solutions in healthcare data engineering and outline the future directions needed to support safe, scalable, and trustworthy AI integration in the United States healthcare system. The intended core contributions of this article are to: (i) identify the need for reliable AI systems for healthcare, (ii) explore challenges associated with implementing AI systems in healthcare from a data engineer’s perspective, and (iii) analyze key limitations of data engineering as it applies to the implementation of AI systems in healthcare. It must be noted that one of the key limitations of this narrative review is that the authors mostly used citations from MDPI journals.

Leveraging Artificial Intelligence (AI) ethically in connected healthcare systems requires a quantifiable framework that measures not only outcome correctness, but also the clarity, auditability, and ethical acceptability of model explanations in high-stakes clinical and cybersecurity workflows. This manuscript first presents a narrative review of ethical risks and countermeasures in Healthcare Internet of Things (HIoT) and explains why existing performance metrics are insufficient for trustworthy deployment. We then formalize a quantitative metric called Ethical Explainability (E_e) as a composite index integrating (1) a Human Agreement Ratio (HAR), capturing concordance between AI recommendations (and their rationale) and a calibrated expert consensus, and (2) an Entropy Reduction Index (ERI), capturing the proportional reduction in expert uncertainty after receiving an explanation, operationalized via probability-elicitation questionnaires mapped to Shannon entropy. Designed for HIoT security monitoring, E_e links transparency with governance-ready evidence of trustworthiness for human–AI collaboration.

Missing data are a persistent challenge in large healthcare datasets, often undermining both statistical validity and machine learning performance when handled using simplistic assumptions. In this work, we examine how entropy-based diagnostics can guide the selection of imputation strategies for high-dimensional health survey data using the National Health and Nutrition Examination Survey (NHANES) 2021–2023. Shannon entropy is used to identify variables with structurally complex missingness, and a range of classical approaches (mean imputation, k-nearest neighbors, and multiple imputation by chained equations) are evaluated alongside deep generative methods, including variational autoencoders, generative adversarial networks (GANs), Wasserstein GANs (WGANs), and diffusion-based models. All methods are compared under a controlled masked-entry evaluation using root mean square error (RMSE) and Kolmogorov–Smirnov (KS) statistics to capture both reconstruction accuracy and distributional fidelity. Results show that diffusion-based models provide the most consistent balance between numerical accuracy and distributional preservation across high-entropy dietary variables, while WGAN demonstrates competitive performance for selected distributions. Structural equation modeling further indicates that entropy is a useful diagnostic signal for identifying variables that are difficult to reconstruct. Overall, this study provides a reproducible framework for aligning imputation strategy with missingness complexity in healthcare data, with implications for improving reliability in downstream analytics.

Accurate and early identification of ocular diseases is essential to prevent vision impairment and enable timely medical intervention. In routine clinical practice, ophthalmologists rely on a structured diagnostic workflow that incorporates multiple imaging modalities to manually assess and diagnose ocular diseases. However, interpreting each modality requires significant clinical experience and can be time-consuming. These limitations can be effectively addressed through the application of AI (Artificial intelligence)-driven multimodal fusion techniques. In this study, we conducted an empirical investigation to assess the impact of different fusion strategies—including early, intermediate, and late fusion—on diagnostic performance, training requirements, and interpretability. The proposed methodology was evaluated using three publicly available datasets: FFA-Fundus (Fundus fluorescein angiography), GAMMA (Glaucoma Analysis and Multi-Modal Assessment), and OLIVES (Ophthalmic Labels to Investigate Visual Eye Semantics). Experimental results demonstrate that multimodal feature fusion improves disease detection performance. Although fused models typically required an increase in training parameters compared to single-modality models, they provided interpretability on par with that of individual single-modal networks. However, inference time increased by approximately 50% for multimodal architectures. These findings underscore the value of integrating diverse ophthalmic imaging modalities to enhance diagnostic accuracy in automated disease detection systems. At the same time, the results highlight that unimodal models containing highly discriminative features can also perform competitively, particularly when a single modality is sufficient for disease identification. Multimodal fusion provides the greatest benefit in scenarios where complementary information across modalities contributes distinct and non-redundant features. Furthermore, fusing all available modalities may not be optimal due to increased computational cost and reduced inference efficiency; thus, selective modality integration and lightweight fusion strategies are essential to balance accuracy, interpretability, and efficiency in clinical deployment.

Background: The digitization of health records was expected to improve data quality and accessibility, yet incompleteness remains a widespread challenge that undermines clinical care, interoperability, and downstream analytics. Problem: Evidence shows that missing and under-recorded elements in electronic health records (EHRs) are largely driven by process gaps across patients, providers, technology, and policy—not solely by technical limitations. Objective: This perspective integrates conceptual foundations of incompleteness, synthesizes cross-country evidence, and examines process-level drivers and consequences, with an emphasis on how missingness propagates bias in AI and machine learning systems. Contribution: We present a unifying taxonomy, highlight complementary approaches (e.g., Record Strength Score, distributional testing, and workflow studies), and we propose a pragmatic agenda for mitigation through technical, organizational, governance, and patient-centered levers. Conclusions: While EHR incompleteness cannot be fully eliminated, it can be systematically mitigated through standards, workflow redesign, patient engagement, and governance—essential steps toward building safe, equitable, and effective learning health systems.

This study explores how patients and stakeholders envision integrated digital health systems. Background/Objectives: Integrating artificial intelligence (AI), wearable data, electronic health records (EHRs), and patient-reported outcomes could enable proactive and personalized healthcare. However, current solutions remain fragmented and poorly aligned with user expectations. This study aimed to explore patient and stakeholder needs for AI-driven integration and propose a conceptual framework to inform future system design. Methods: As part of the NSF Innovation Corps (I-Corps) program, we conducted semi-structured interviews with 44 participants representing Health Enthusiasts, Chronic Condition Managers, and Low-Engagement Users. Interviews followed the I-Corps customer discovery framework and were thematically analyzed using a hybrid deductive–inductive approach. Results: Participants highlighted four priorities: (i) interoperability and unification of data from wearables, EHRs, and self-reports; (ii) actionable personalization with predictive insights; (iii) trust and transparency in AI recommendations, often requiring clinician oversight; and (iv) usability through low-friction, intuitive interfaces. Age- and persona-specific differences emerged: younger participants favoring predictive features and older participants emphasizing safety, reassurance, and clinical integration. Conclusions: This exploratory qualitative study identified stakeholder needs that informed a conceptual framework for integrated digital health platforms. While preliminary, the framework provides a blueprint for future technical development and validation of patient- and provider-centered systems.

Dental care is a well-established marker of both oral and systemic health, driven by behavioral, socioeconomic, and geographic factors. This study aimed to develop and evaluate machine learning models to classify the presence and severity of permanent tooth loss in U.S. adults using the 2022 Behavioral Risk Factor Surveillance System (BRFSS) dataset. We analyzed responses from 365,803 adults after recoding demographic, behavioral, socioeconomic, and access variables. Ten supervised classifiers were trained and evaluated using stratified 80/20 train–test splits, with ANOVA-based selection for the binary task and Pearson correlation plus engineered features for the multiclass task. Performance was assessed by accuracy, AUC, precision, recall, and specificity. For binary classification (any loss vs. none), XGBoost achieved the highest performance (AUC = 0.786; accuracy = 71.4%), with CatBoost close behind (AUC = 0.711). For multiclass severity (none, 1–5, 6+, all teeth removed), an ensemble of gradient-boosting models achieved strong discrimination (macro-AUC = 0.752). Key predictors included age, smoking, education, income, and general health. These findings demonstrate that large-scale survey–based ML models can support oral health surveillance by identifying high-risk groups and informing targeted prevention strategies.