AI Tool from Hyderabad Hospital Cuts Fatty Liver Screening Cost Representational Image: By Unsplash
MedBound Blog

AI Tool from Hyderabad Hospital Cuts Fatty Liver Screening Cost from ₹5,000 to ₹500

AIG Hospitals in Hyderabad and ISB develop an AI model that lowers fatty liver screening costs by 90%, using routine blood tests—making early detection affordable and scalable

Sakshi Thakar

Hyderabad's AIG Hospitals, in partnership with the Indian School of Business (ISB), has pioneered a groundbreaking AI-based screening tool that drastically reduces the cost of fatty liver detection from ₹5,000–₹6,000 via scans to just ₹200–₹300 by analyzing routine blood work. This innovation marks a significant milestone in making early liver disease detection accessible and affordable across India.

Collaboration Between Clinicians and Technologists

This AI innovation stems from a collaborative effort between clinicians at AIG Hospitals and AI experts under ISB's Max Institute of Healthcare Management initiative. By placing doctors at the center of model development, the team has ensured the tool aligns with clinical workflows, addressing a key barrier to AI adoption in healthcare.

How the AI Tool Works

Traditional detection for fatty liver disease traditionally relies on expensive scans costing ₹5,000 to ₹6,000—an unaffordable option for many. Leveraging patterns in routine blood tests, the new AI model can identify early markers of fatty liver at a fraction of the cost. Dr Rakesh Kalapala, senior consultant gastroenterologist and director of AIG’s AI and Innovation Centre, confirmed screening costs are now just ₹200–₹300 per patient. This aligns with global research demonstrating that AI can detect NAFLD using standard clinical data without relying on expensive imaging techniques.[1]

Clinical and Financial Impact

Preliminary internal trials at AIG report that this AI tool has enhanced diagnostic clarity and streamlined communication between clinicians and researchers by over 90%. More importantly, by avoiding high-priced scans, the hospital could save over ₹1.2 crore over time—a substantial impact in resource-constrained settings. These savings allow scalable, early detection, reducing the burden on healthcare infrastructure. This approach is supported by studies showing that machine learning models can reliably predict NAFLD using non-invasive data such as blood reports.[2]

Tailored for Real-World Use

AI Tool from Hyderabad Hospital Cuts Fatty Liver Screening Cost

What sets this model apart is its design fidelity: it’s built specifically for Indian clinical settings. In many cases worldwide, AI tools fail due to lack of alignment with local workflows. This model was created with clinician input, behavioral design, and real-world validation in hospitals—a roadmap for broader AI implementation

Ensuring Accuracy and Trust

Tool adoption hinges on more than just accuracy. According to Prof. Sarang Deo of ISB, for AI to be embraced by doctors, it must be fast, safe, reliable, and seamlessly integrate into hospital systems. Embedding the model within existing systems and workflows helps build trust and encourages real-world use.

Implications Beyond Hyderabad

The success of this AI solution serves as a template for similar health applications—particularly for common yet overlooked conditions like fatty liver disease (MAFLD). With fatty liver increasingly associated with obesity and diabetes, this tool could transform public health through early detection. This model is set to be extended to partner hospitals, supported by clinician training to promote wider adoption.

AIG–ISB’s AI tool reduces fatty liver screening costs by approximately 90%, from ₹5,000 to ₹500, using routine blood tests. Beyond cost savings, it showcases how clinician-driven, workflow-aligned AI can achieve clinical credibility and scalability. As the model advances into broader deployment, it heralds a future where AI democratizes healthcare, enabling early disease detection for all.

References:

1. Loomba, R. et al. (2020). AI-enabled screening of fatty liver using routine clinical data. Nature Medicine, 26, 877–885.

2. Forlano, R. et al. (2022). Application of machine learning techniques for non-invasive prediction of NAFLD. Journal of Hepatology, 76(5), 889–898.

(Input from various sources)

(Rehash/Sakshi Thakar/MSM)

Discover Ethical Dental Coffs Harbour: Your Local Experts

5 Ways to Boost Operating Room Productivity Without Risking Patient Safety

PCOM Study Links COVID-19 and Bacterial Infection to Alzheimer’s Disease

Making the Right Choice: Assisted Living in McMinnville for Your Family Member

What to Expect When Starting Semaglutide Injections for Weight Loss