This project explores Hepatitis C as a major public health concern and examines how artificial intelligence can support healthcare systems in improving prevention, screening, and treatment outcomes.
A comprehensive overview of HCV — from what it is and how it spreads, to how it is diagnosed, prevented, and treated.
Hepatitis C is a bloodborne viral infection caused by the hepatitis C virus (HCV) that primarily targets the liver, often leading to chronic disease if untreated. It can be acute (short-term) or chronic (long-term), and is a major global health concern due to the risk of progressive liver damage.
Advances in direct-acting antiviral (DAA) therapies have made hepatitis C highly curable, with cure rates exceeding 95%, significantly reducing long-term complications. There is currently no widely available vaccine. (WHO, 2025; CDC, 2025)
HCV is primarily transmitted through exposure to infected blood. In high-income countries, most infections are due to shared needles among injection drug users. In low-to-middle-income countries, unsafe medical practices are the primary driver. (WHO, 2025)
HCV infection is often asymptomatic in its early stages, making routine screening critical. (CDC, 2025)
Because of this silent progression, early detection through routine screening is essential to prevent liver failure and hepatocellular carcinoma.
HCV infection is diagnosed through a two-step process. Routine screening is recommended for all adults at least once in their lifetime. (CDC, 2025; WHO, 2025)
There is currently no vaccine for HCV. Prevention focuses on reducing exposure to infected blood. (WHO, 2025; CDC, 2025)
Treatment of HCV has been transformed by direct-acting antiviral (DAA) medications, which cure more than 95% of cases in just 8–12 weeks of oral therapy. (WHO, 2025)
References: Bernal & Soti (2023); WHO (2025); CDC (2025)
HCV is curable in 95% of cases, yet millions remain undiagnosed. The barrier is not the cure — it is detection, access, and timely clinical decisions. AI is closing that gap.
3–4 million new HCV infections occur globally every year. Most infected individuals show no symptoms for months or years, leaving a large portion of the population undiagnosed.
AI scans electronic health records to proactively flag individuals with known HCV risk factors — IV drug use, blood transfusions, unsafe practices — before they ever show symptoms. (Bal, 2024)
A 5-item AI risk score assessing gender, place of origin, IV drug use, self-perceived risk, and past unexplained liver disease achieved 98% sensitivity (NLR: 0.05) — a cost-effective alternative to universal screening. (Martínez-Sanz et al., 2022)
Traditional liver biopsy carries a 0.01% mortality risk and is subjective — METAVIR inter-observer agreement is only moderate (κ = 0.58), meaning two pathologists can grade the same biopsy differently. (Chowdhury & Mehta, 2023)
An AI liver function test system achieved >90% diagnostic accuracy and improved detection by 43% vs. standard care — using routine lab samples with no biopsy required. (Bal, 2024)
The deep learning DLRE model predicted cirrhosis (F4) at 97% accuracy and advanced fibrosis at 98% accuracy. AI ultrasound achieved 92% accuracy vs. 76.1% for experienced radiologists. (Bal, 2024)
The Intelligent Hepatitis C Stage Diagnosis System (ANN model) staged HCV fibrosis at 94.44% precision using 18 clinical features — outperforming SVM (85.6%), Random Forest (90.3%), and XGBoost (81%). No biopsy needed. (Butt et al., 2021)
A deep learning RNN trained on 48,000 patients with HCV-related cirrhosis predicted liver cancer risk at 75% accuracy vs. 68% for traditional models — capturing long-term patterns traditional models miss. (Ioannou et al., 2020)
A CNN trained on ECG recordings from 5,212 patients detected cirrhosis at 90% accuracy (sensitivity 84.9%, specificity 83.2%) — using only ECG data, no biopsy or imaging required. (Bal, 2024)
| Measure | Traditional Liver Biopsy | AI-Based System |
|---|---|---|
| Invasiveness | Invasive (needle insertion) | Non-invasive |
| Mortality risk | ~0.01% | None |
| Observer agreement | Moderate (κ = 0.58) | Consistent, reproducible |
| Cirrhosis accuracy | Variable by pathologist | 97% (DLRE model) |
| Advanced fibrosis accuracy | Variable by pathologist | 98% (DLRE model) |
| Detection rate vs. standard care | Baseline | +43% improvement |
| Cost | High (procedure + lab) | Uses existing EHR/lab data |
Sources: Chowdhury & Mehta (2023); Bal (2024); Wang et al. (2019); Butt et al. (2021)
The use of AI in healthcare requires clear regulations to ensure safety, accuracy, and ethical implementation. Here are the key frameworks governing AI-powered HCV care.
Despite major advancements in screening, treatment, and technology, hepatitis C continues to be a persistent public health challenge. Artificial intelligence has the potential to significantly improve early detection by analyzing electronic health records and identifying high-risk individuals. However, the effectiveness of AI is directly shaped by the policies and laws that govern healthcare data.
Privacy regulations are essential for protecting sensitive patient information, but they can also limit how data is shared. At the same time, laws like the 21st Century Cures Act aim to improve data access, helping create better conditions for AI to function effectively. The challenge is balancing innovation with patient privacy — and ensuring AI-driven solutions benefit all populations equitably.
HIPAA is the primary law protecting patient health information (PHI) in the United States. It applies to hospitals, clinics, insurance companies, and any organization that handles protected health information.
The law requires that patient data is stored securely, shared only when necessary, and accessed only by authorized individuals. Organizations using AI must implement safeguards like encryption, secure databases, and strict access controls to remain compliant.
42 CFR Part 2 provides additional privacy protections specifically for individuals receiving substance use disorder (SUD) treatment. It restricts how information related to substance use can be disclosed, requiring explicit patient consent before sharing.
This helps build trust and encourages individuals to seek care without fear of stigma or exposure.
The CDC recommends screening all adults 18+ at least once in their lifetime, with routine screening for high-risk individuals and pregnant women during each pregnancy.
The USPSTF recommends screening for adults aged 18–79, which directly influences insurance coverage. Preventive services following strong USPSTF recommendations are typically covered without cost to the patient.
The ACA requires most private insurance plans to cover preventive services with strong clinical recommendations — such as hepatitis C screening — without charging copays or deductibles.
This law plays a major role in making prevention financially accessible, especially for vulnerable populations who are at higher risk of hepatitis C.
The 21st Century Cures Act focuses on improving healthcare innovation and access to information. A key component prevents "information blocking" — healthcare providers cannot unnecessarily restrict access to electronic health data.
This also promotes transparency by allowing patients to access their own health information, increasing engagement in their care.
The FDA regulates medical devices, including certain AI-based tools used in healthcare. If an AI system assists with diagnosis, risk prediction, or treatment decisions, it may need to go through FDA review to ensure it is safe and effective.
FDA oversight helps ensure AI tools meet quality standards, reduce the risk of inaccurate predictions, minimize algorithmic bias, and provide reliable support to healthcare providers.
Curated videos and materials to deepen your understanding of Hepatitis C and how AI is transforming its prevention and care.
Graduate students at San José State University combining expertise in healthcare, technology, and policy to make complex health information accessible and actionable.
As graduate students in Health Informatics, our goal is to make complex health information accessible and actionable. By combining expertise in healthcare, technology, and policy, we hope this site serves as a meaningful resource for anyone seeking to understand Hepatitis C and the role artificial intelligence is playing in transforming how it is detected, treated, and ultimately eliminated.
We began by identifying Hepatitis C as a topic where technology and public health intersect in meaningful ways. We drew on each member's academic and professional background to approach the topic from multiple angles, holding collaborative sessions to align our research, critique each other's findings, and ensure the final product meets a high standard of rigor and accuracy.
Holds a Bachelor of Science in Computer Science and currently works for a government agency with experience in public sector project management and data analysis. Responsible for researching and developing the policies, laws, and clinical recommendations section, ensuring information is accurate, clear, and relevant to hepatitis C prevention and the use of AI in healthcare.
A first-year Health Informatics student holding a Bachelor of Science in Business Management, with a professional background spanning Project Management, Operations, and UX Research. For this project, Dani led data analysis and evaluated AI methodologies across hepatitis C surveillance, covering prevention, screening, and treatment outcomes. She also designed the presentation template and website, and guided the team's collaborative process from start to finish.
Holds a Bachelor of Science in Biology from CSU Long Beach. Has worked for Los Angeles County's Department of Health Services and LA County's Department of Public Health in medical laboratories and disease intervention. First-year student in the Health Informatics MS program. Responsible for the clinical background and treatment evolution section.