Journal of Advanced Scientific Research

National-Scale Predictive Analytics for Medicare and Medicaid: An AI-Driven Approach to Identifying High-Risk Populations and Reducing Healthcare Costs

Tan Tho Nguyen — 2026-06-22

Rising expenditures in Medicare and Medicaid continue to challenge the long-term sustainability of public healthcare financing in the United
States. A relatively small proportion of beneficiaries accounts for a disproportionate share of annual spending due to chronic disease burden,
repeated hospitalizations, fragmented care pathways, and unmet social needs. Early identification of these high-risk and high-cost populations is
therefore essential for improving outcomes while controlling avoidable expenditure. This study examines the application of artificial intelligencedriven
predictive analytics at national scale to strengthen population health management across Medicare and Medicaid programs. Using
integrated claims records, electronic health records, demographic indicators, utilization histories, and selected social determinants of health
variables, multiple machine learning models were developed to estimate future hospitalization risk, readmission probability, and annual cost
escalation. Comparative evaluation indicates that ensemble and deep learning approaches outperform conventional regression-based methods
in risk stratification accuracy, sensitivity, and cost forecasting performance. Prior utilization, multimorbidity burden, medication complexity,
emergency department use, and socioeconomic vulnerability emerged as the most influential predictors. Simulated deployment results suggest
that earlier targeting of case management, preventive outreach, and transitional care programs could reduce unnecessary admissions and moderate
total program spending. The findings demonstrate that scalable predictive systems can support more proactive and efficient allocation of limited
healthcare resources. From a policy perspective, national implementation of AI-enabled analytics may improve care coordination, strengthen
value-based purchasing strategies, and enhance equity by identifying underserved beneficiaries with elevated risk profiles.

Nationwide Multimodal Artificial Intelligence Framework for Early Prediction of Chronic Disease Progression Using Electronic Health Records and Social Determinants of Health

Trang Huynh — 2026-06-22

Chronic disease progression remains a major challenge for national healthcare systems because risk often develops gradually across clinical,
behavioral, environmental, and socioeconomic dimensions. Existing prediction models frequently rely on limited electronic health record
variables and may overlook unstructured clinical notes, longitudinal patient trajectories, physiologic signals, and social determinants of health
that influence disease worsening. This paper proposes a nationwide multimodal artificial intelligence framework for early prediction of chronic
disease progression by integrating structured EHR data, clinical narratives, laboratory histories, medication records, physiologic indicators,
and SDOH variables. The framework combines deep learning, transformer-based EHR modeling, natural language processing, multimodal
fusion, explainable AI, and fairness auditing to support early risk identification and patient stratification across diverse healthcare settings. It
also emphasizes model validation across demographic, geographic, and socioeconomic groups to reduce algorithmic bias and improve clinical
reliability. The proposed framework offers a scalable pathway for preventive intervention, chronic care planning, population health surveillance,
and equitable clinical decision support. By combining medical and social risk signals, the study contributes to a more comprehensive and nationally
deployable approach for predicting chronic disease progression before severe complications occur.

A Study of Clinico-radiological, CSF Profile and Outcome of Scrub Typhus Meningoencephalitis with Special Reference to Atypical Presentation

Abu Khair — 2026-06-23

Background: Scrub typhus is an endemic disease, which is caused by Orientia tsutsugamushi. Scrub typhus affects a region known as “the
tsutsugamushi triangle”, which starts from the northern part of Japan and extends to the southern part of Northern Australia, northern part of
Russia, and Pakistan. The present study focued to generate evidence regarding clinical manifestation and treatment outcome of the scrub typhus
meningoencephalitis cases admitted in a tertiary care centre of Burdwan, West Bengal.
Materials & Methods: The cross-sectional observational study was done at inpatient department of General Medicine, Burdwan Medical College
& Hospital, Burdwan, West Bengal. The duration of study was 18 months. Diagnosed patients of scrub typhus meningoencephalitis, attending
inpatient (IPD) department of General Medicine were included in study population. Calculated final sample size was 68. Liver function test,
renal function test, coagulation profile, CSF analysis, NCCT brain, MRI brain was done.
Results: Prevalence of serious scrub typhus infection was 68.6%. Nearly 44.3% study subjects were within 18-36 years age group; majority were
male (64.3%). 70% had shown Eschar formation. 44.3% had comorbidities. Nearly 43% study subjects presented with pallor, 38% presented
with icterus, 27.6% with hypotension, 21% with hepatomegaly, 18.4% with splenomegaly and 22% with neck rigidity. Nearly 57.1% study
subjects had bleeding disorder, 37.1% with hepatitis, 25.7% with acute kidney injury and 22.9% with meningitis. Age and Eschar formation had
significant associations with disease severity of study subjects. Coagulation profile (platelet, INR, D-dimer, APTT), serum urea, creatinine, liver
function (bilirubin, SGOT, SGPT, albumin) and total count of WBC, Hemoglobin, CRP and serum ferritin were also significantly associated
with treatment outcome.
Conclusion: Atypical presentations in scrub typhus were common and they were needed to be treated early to reduce morbidity & mortality
burden of the disease.

Clinical Profile of AnoRectal Bleeding in Patients Attending a Tertiary Care Surgical Unit – A cross-sectional study from Northeast India

Bhupendra Kumar Sahu — 2026-06-23

Background: Anorectal bleeding is a common surgical presentation, often linked to benign conditions such as haemorrhoids and fissures, but
occasionally indicative of malignancy. Regional data from Northeast India remains limited.
Objective: To evaluate the sociodemographic profile, clinical presentation, and etiological spectrum of patients presenting with anorectal
bleeding at a tertiary care surgical unit.
Methods: A crosssectional study was conducted at AGMC & GBP Hospital, Agartala, enrolling 629 consecutive patients with rectal bleeding
between June 2024 and December 2025. Data on demographics, clinical features, comorbidities, digital rectal examination findings, haemoglobin
levels, and complications were collected using a structured proforma and analysed descriptively.
Results: The majority of patients were aged 40–60 years (37.2%) and male (69.8%). Urban residents comprised 59.9% of the cohort. Pain
(42.9%), constipation (32.8%), and rectal mass (24.3%) were the leading symptoms. Digital rectal examination revealed anal fissure (47.7%)
and haemorrhoids (46.3%) as the predominant causes, while malignancy was detected in 1.9%. Anaemia was the most frequent complication
(13.5%). Most patients had haemoglobin >10 g/dl (71.4%).
Conclusion: Anorectal bleeding in this region is primarily attributable to benign anorectal disorders, though malignancy, though infrequent,
remains a critical differential diagnosis. Early evaluation and lifestyle modification are essential to reduce morbidity and improve outcomes.